diff --git a/providers/maxlayer/logo.svg b/providers/maxlayer/logo.svg
new file mode 100644
index 00000000000..409d3c9f08c
--- /dev/null
+++ b/providers/maxlayer/logo.svg
@@ -0,0 +1,5 @@
+<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 32 32" fill="currentColor">
+  <path d="M16 3.5 29 11 16 18.5 3 11z" />
+  <path d="M4.5 15.6 16 22.2l11.5-6.6" fill="none" stroke="currentColor" stroke-width="2.2" stroke-linecap="round" stroke-linejoin="round" opacity="0.65" />
+  <path d="M4.5 20.6 16 27.2l11.5-6.6" fill="none" stroke="currentColor" stroke-width="2.2" stroke-linecap="round" stroke-linejoin="round" opacity="0.3" />
+</svg>
diff --git a/providers/maxlayer/models/anthropic/claude-fable-5.toml b/providers/maxlayer/models/anthropic/claude-fable-5.toml
new file mode 100644
index 00000000000..cca06f6fc35
--- /dev/null
+++ b/providers/maxlayer/models/anthropic/claude-fable-5.toml
@@ -0,0 +1,12 @@
+base_model = "anthropic/claude-fable-5"
+structured_output = true
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 11.394
+output = 56.97
+cache_read = 1.1394
+cache_write = 14.2425
diff --git a/providers/maxlayer/models/anthropic/claude-haiku-4.5.toml b/providers/maxlayer/models/anthropic/claude-haiku-4.5.toml
new file mode 100644
index 00000000000..1e21c695073
--- /dev/null
+++ b/providers/maxlayer/models/anthropic/claude-haiku-4.5.toml
@@ -0,0 +1,19 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "anthropic/claude-haiku-4-5"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 1_024
+max = 63_999
+
+[cost]
+input = 1.1394
+output = 5.697
+cache_read = 0.11394
+cache_write = 1.42425
diff --git a/providers/maxlayer/models/anthropic/claude-opus-4.1.toml b/providers/maxlayer/models/anthropic/claude-opus-4.1.toml
new file mode 100644
index 00000000000..2d2155c8545
--- /dev/null
+++ b/providers/maxlayer/models/anthropic/claude-opus-4.1.toml
@@ -0,0 +1,19 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "anthropic/claude-opus-4-1"
+structured_output = false
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 1_024
+max = 31_999
+
+[cost]
+input = 17.091
+output = 85.455
+cache_read = 1.7091
+cache_write = 21.36375
diff --git a/providers/maxlayer/models/anthropic/claude-opus-4.5.toml b/providers/maxlayer/models/anthropic/claude-opus-4.5.toml
new file mode 100644
index 00000000000..a08fd9c08b0
--- /dev/null
+++ b/providers/maxlayer/models/anthropic/claude-opus-4.5.toml
@@ -0,0 +1,23 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "anthropic/claude-opus-4-5"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 1_024
+max = 63_999
+
+[cost]
+input = 5.697
+output = 28.485
+cache_read = 0.5697
+cache_write = 7.12125
diff --git a/providers/maxlayer/models/anthropic/claude-opus-4.6.toml b/providers/maxlayer/models/anthropic/claude-opus-4.6.toml
new file mode 100644
index 00000000000..cc09548d690
--- /dev/null
+++ b/providers/maxlayer/models/anthropic/claude-opus-4.6.toml
@@ -0,0 +1,21 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "anthropic/claude-opus-4-6"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "max"]
+
+[[reasoning_options]]
+type = "budget_tokens"
+
+[cost]
+input = 5.697
+output = 28.485
+cache_read = 0.5697
+cache_write = 7.12125
diff --git a/providers/maxlayer/models/anthropic/claude-opus-4.7-fast.toml b/providers/maxlayer/models/anthropic/claude-opus-4.7-fast.toml
new file mode 100644
index 00000000000..e92afa2b4a5
--- /dev/null
+++ b/providers/maxlayer/models/anthropic/claude-opus-4.7-fast.toml
@@ -0,0 +1,19 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "anthropic/claude-opus-4-7"
+name = "Claude Opus 4.7 (Fast)"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 34.182
+output = 170.91
+cache_read = 3.4182
+cache_write = 42.7275
diff --git a/providers/maxlayer/models/anthropic/claude-opus-4.7.toml b/providers/maxlayer/models/anthropic/claude-opus-4.7.toml
new file mode 100644
index 00000000000..3bca5eede3c
--- /dev/null
+++ b/providers/maxlayer/models/anthropic/claude-opus-4.7.toml
@@ -0,0 +1,18 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "anthropic/claude-opus-4-7"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 5.697
+output = 28.485
+cache_read = 0.5697
+cache_write = 7.12125
diff --git a/providers/maxlayer/models/anthropic/claude-opus-4.8-fast.toml b/providers/maxlayer/models/anthropic/claude-opus-4.8-fast.toml
new file mode 100644
index 00000000000..4da3b33b5a6
--- /dev/null
+++ b/providers/maxlayer/models/anthropic/claude-opus-4.8-fast.toml
@@ -0,0 +1,19 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "anthropic/claude-opus-4-8"
+name = "Claude Opus 4.8 (Fast)"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 11.394
+output = 56.97
+cache_read = 1.1394
+cache_write = 14.2425
diff --git a/providers/maxlayer/models/anthropic/claude-opus-4.8.toml b/providers/maxlayer/models/anthropic/claude-opus-4.8.toml
new file mode 100644
index 00000000000..fa63917a08e
--- /dev/null
+++ b/providers/maxlayer/models/anthropic/claude-opus-4.8.toml
@@ -0,0 +1,19 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "anthropic/claude-opus-4-8"
+temperature = true
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 5.697
+output = 28.485
+cache_read = 0.5697
+cache_write = 7.12125
diff --git a/providers/maxlayer/models/anthropic/claude-opus-5-fast.toml b/providers/maxlayer/models/anthropic/claude-opus-5-fast.toml
new file mode 100644
index 00000000000..85672524540
--- /dev/null
+++ b/providers/maxlayer/models/anthropic/claude-opus-5-fast.toml
@@ -0,0 +1,19 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "anthropic/claude-opus-5"
+name = "Claude Opus 5 (Fast)"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 11.394
+output = 56.97
+cache_read = 1.1394
+cache_write = 14.2425
diff --git a/providers/maxlayer/models/anthropic/claude-opus-5.toml b/providers/maxlayer/models/anthropic/claude-opus-5.toml
new file mode 100644
index 00000000000..27a148b8464
--- /dev/null
+++ b/providers/maxlayer/models/anthropic/claude-opus-5.toml
@@ -0,0 +1,19 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "anthropic/claude-opus-5"
+temperature = true
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 5.697
+output = 28.485
+cache_read = 0.5697
+cache_write = 7.12125
diff --git a/providers/maxlayer/models/anthropic/claude-sonnet-4.5.toml b/providers/maxlayer/models/anthropic/claude-sonnet-4.5.toml
new file mode 100644
index 00000000000..d95b5fdb622
--- /dev/null
+++ b/providers/maxlayer/models/anthropic/claude-sonnet-4.5.toml
@@ -0,0 +1,22 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "anthropic/claude-sonnet-4-5"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 1_024
+max = 63_999
+
+[cost]
+input = 3.4182
+output = 17.091
+cache_read = 0.34182
+cache_write = 4.27275
+
+[limit]
+context = 1_000_000
diff --git a/providers/maxlayer/models/anthropic/claude-sonnet-4.6.toml b/providers/maxlayer/models/anthropic/claude-sonnet-4.6.toml
new file mode 100644
index 00000000000..4e3c6e50311
--- /dev/null
+++ b/providers/maxlayer/models/anthropic/claude-sonnet-4.6.toml
@@ -0,0 +1,21 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "anthropic/claude-sonnet-4-6"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "max"]
+
+[cost]
+input = 3.4182
+output = 17.091
+cache_read = 0.34182
+cache_write = 4.27275
+
+[limit]
+output = 128_000
diff --git a/providers/maxlayer/models/anthropic/claude-sonnet-5.toml b/providers/maxlayer/models/anthropic/claude-sonnet-5.toml
new file mode 100644
index 00000000000..bb4143425fc
--- /dev/null
+++ b/providers/maxlayer/models/anthropic/claude-sonnet-5.toml
@@ -0,0 +1,18 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "anthropic/claude-sonnet-5"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 2.2788
+output = 11.394
+cache_read = 0.22788
+cache_write = 2.8485
diff --git a/providers/maxlayer/models/bytedance-seed/seed-2.0-code.toml b/providers/maxlayer/models/bytedance-seed/seed-2.0-code.toml
new file mode 100644
index 00000000000..d611e50cfa9
--- /dev/null
+++ b/providers/maxlayer/models/bytedance-seed/seed-2.0-code.toml
@@ -0,0 +1,15 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "bytedance-seed/seed-2.0-code"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 0.5697
+output = 3.4182
diff --git a/providers/maxlayer/models/bytedance-seed/seed-2.0-lite.toml b/providers/maxlayer/models/bytedance-seed/seed-2.0-lite.toml
new file mode 100644
index 00000000000..9da9ccd698d
--- /dev/null
+++ b/providers/maxlayer/models/bytedance-seed/seed-2.0-lite.toml
@@ -0,0 +1,20 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "bytedance-seed/seed-2.0-lite"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 0.28485
+output = 2.2788
+
+[limit]
+context = 262_144
+output = 131_072
diff --git a/providers/maxlayer/models/bytedance-seed/seed-2.0-mini.toml b/providers/maxlayer/models/bytedance-seed/seed-2.0-mini.toml
new file mode 100644
index 00000000000..c2728c614ec
--- /dev/null
+++ b/providers/maxlayer/models/bytedance-seed/seed-2.0-mini.toml
@@ -0,0 +1,20 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "bytedance-seed/seed-2.0-mini"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 0.11394
+output = 0.45576
+
+[limit]
+context = 262_144
+output = 131_072
diff --git a/providers/maxlayer/models/cohere/command-r-08-2024.toml b/providers/maxlayer/models/cohere/command-r-08-2024.toml
new file mode 100644
index 00000000000..0d7491377d2
--- /dev/null
+++ b/providers/maxlayer/models/cohere/command-r-08-2024.toml
@@ -0,0 +1,6 @@
+base_model = "cohere/command-r-08-2024"
+structured_output = true
+
+[cost]
+input = 0.17091
+output = 0.68364
diff --git a/providers/maxlayer/models/cohere/command-r-plus-08-2024.toml b/providers/maxlayer/models/cohere/command-r-plus-08-2024.toml
new file mode 100644
index 00000000000..b57cc0e737a
--- /dev/null
+++ b/providers/maxlayer/models/cohere/command-r-plus-08-2024.toml
@@ -0,0 +1,6 @@
+base_model = "cohere/command-r-plus-08-2024"
+structured_output = true
+
+[cost]
+input = 2.8485
+output = 11.394
diff --git a/providers/maxlayer/models/cohere/command-r7b-12-2024.toml b/providers/maxlayer/models/cohere/command-r7b-12-2024.toml
new file mode 100644
index 00000000000..9fcaa865746
--- /dev/null
+++ b/providers/maxlayer/models/cohere/command-r7b-12-2024.toml
@@ -0,0 +1,7 @@
+base_model = "cohere/command-r7b-12-2024"
+tool_call = false
+structured_output = true
+
+[cost]
+input = 0.042729
+output = 0.17091
diff --git a/providers/maxlayer/models/deepseek/deepseek-chat.toml b/providers/maxlayer/models/deepseek/deepseek-chat.toml
new file mode 100644
index 00000000000..007d73e117e
--- /dev/null
+++ b/providers/maxlayer/models/deepseek/deepseek-chat.toml
@@ -0,0 +1,11 @@
+base_model = "deepseek/deepseek-chat"
+attachment = false
+structured_output = true
+
+[cost]
+input = 0.293282
+output = 1.172102
+
+[limit]
+context = 163_840
+output = 16_000
diff --git a/providers/maxlayer/models/deepseek/deepseek-r1.toml b/providers/maxlayer/models/deepseek/deepseek-r1.toml
new file mode 100644
index 00000000000..60e623c7a9f
--- /dev/null
+++ b/providers/maxlayer/models/deepseek/deepseek-r1.toml
@@ -0,0 +1,11 @@
+base_model = "deepseek/deepseek-r1"
+structured_output = true
+reasoning_options = []
+
+[cost]
+input = 0.79758
+output = 2.8485
+
+[limit]
+context = 64_000
+output = 16_000
diff --git a/providers/maxlayer/models/deepseek/deepseek-v3.2.toml b/providers/maxlayer/models/deepseek/deepseek-v3.2.toml
new file mode 100644
index 00000000000..e7cd526bbf5
--- /dev/null
+++ b/providers/maxlayer/models/deepseek/deepseek-v3.2.toml
@@ -0,0 +1,16 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "deepseek/deepseek-v3.2"
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.306499
+output = 0.45576
+cache_read = 0.15325
+
+[limit]
+context = 163_840
+output = 65_536
diff --git a/providers/maxlayer/models/deepseek/deepseek-v4-flash-0731.toml b/providers/maxlayer/models/deepseek/deepseek-v4-flash-0731.toml
new file mode 100644
index 00000000000..e7d96c398fe
--- /dev/null
+++ b/providers/maxlayer/models/deepseek/deepseek-v4-flash-0731.toml
@@ -0,0 +1,20 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "deepseek/deepseek-v4-flash-0731"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "high", "max"]
+
+[cost]
+input = 0.159516
+output = 0.319032
+cache_read = 0.031904
+
+[limit]
+context = 1_310_720
+output = 393_216
diff --git a/providers/maxlayer/models/deepseek/deepseek-v4-flash.toml b/providers/maxlayer/models/deepseek/deepseek-v4-flash.toml
new file mode 100644
index 00000000000..882331e365a
--- /dev/null
+++ b/providers/maxlayer/models/deepseek/deepseek-v4-flash.toml
@@ -0,0 +1,22 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "deepseek/deepseek-v4-flash"
+
+[interleaved]
+field = "reasoning_content"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["high", "xhigh"]
+
+[cost]
+input = 0.094115
+output = 0.188229
+cache_read = 0.018824
+
+[limit]
+context = 1_048_576
diff --git a/providers/maxlayer/models/deepseek/deepseek-v4-pro-0813.toml b/providers/maxlayer/models/deepseek/deepseek-v4-pro-0813.toml
new file mode 100644
index 00000000000..16c618d0684
--- /dev/null
+++ b/providers/maxlayer/models/deepseek/deepseek-v4-pro-0813.toml
@@ -0,0 +1,19 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "deepseek/deepseek-v4-pro-0813"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "high", "max"]
+
+[cost]
+input = 1.353608
+output = 4.060822
+cache_read = 0.045121
+
+[limit]
+context = 1_048_576
diff --git a/providers/maxlayer/models/deepseek/deepseek-v4-pro.toml b/providers/maxlayer/models/deepseek/deepseek-v4-pro.toml
new file mode 100644
index 00000000000..fc3869825de
--- /dev/null
+++ b/providers/maxlayer/models/deepseek/deepseek-v4-pro.toml
@@ -0,0 +1,23 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "deepseek/deepseek-v4-pro"
+
+[interleaved]
+field = "reasoning_content"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["high", "xhigh"]
+
+[cost]
+input = 1.82304
+output = 3.64608
+cache_read = 0.153819
+
+[limit]
+context = 1_048_576
+output = 393_216
diff --git a/providers/maxlayer/models/google/gemini-2.5-flash-lite.toml b/providers/maxlayer/models/google/gemini-2.5-flash-lite.toml
new file mode 100644
index 00000000000..72f8de33c4a
--- /dev/null
+++ b/providers/maxlayer/models/google/gemini-2.5-flash-lite.toml
@@ -0,0 +1,21 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "google/gemini-2.5-flash-lite"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 512
+max = 24_576
+
+[cost]
+input = 0.11394
+output = 0.45576
+cache_read = 0.011394
+cache_write = 0.094951
+
+[limit]
+output = 65_535
diff --git a/providers/maxlayer/models/google/gemini-2.5-flash.toml b/providers/maxlayer/models/google/gemini-2.5-flash.toml
new file mode 100644
index 00000000000..d385ceeb5c3
--- /dev/null
+++ b/providers/maxlayer/models/google/gemini-2.5-flash.toml
@@ -0,0 +1,21 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "google/gemini-2.5-flash"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 0
+max = 24_576
+
+[cost]
+input = 0.34182
+output = 2.8485
+cache_read = 0.034182
+cache_write = 0.094951
+
+[limit]
+output = 65_535
diff --git a/providers/maxlayer/models/google/gemini-2.5-pro.toml b/providers/maxlayer/models/google/gemini-2.5-pro.toml
new file mode 100644
index 00000000000..457d87ffaaa
--- /dev/null
+++ b/providers/maxlayer/models/google/gemini-2.5-pro.toml
@@ -0,0 +1,12 @@
+base_model = "google/gemini-2.5-pro"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 128
+max = 32_768
+
+[cost]
+input = 1.42425
+output = 11.394
+cache_read = 0.142425
+cache_write = 0.427275
diff --git a/providers/maxlayer/models/google/gemini-3-flash-preview.toml b/providers/maxlayer/models/google/gemini-3-flash-preview.toml
new file mode 100644
index 00000000000..dd9e12996ea
--- /dev/null
+++ b/providers/maxlayer/models/google/gemini-3-flash-preview.toml
@@ -0,0 +1,20 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "google/gemini-3-flash-preview"
+
+[interleaved]
+field = "reasoning_details"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 0.5697
+output = 3.4182
+cache_read = 0.05697
+cache_write = 0.094951
diff --git a/providers/maxlayer/models/google/gemini-3.1-flash-lite-preview.toml b/providers/maxlayer/models/google/gemini-3.1-flash-lite-preview.toml
new file mode 100644
index 00000000000..a7a5805d5f8
--- /dev/null
+++ b/providers/maxlayer/models/google/gemini-3.1-flash-lite-preview.toml
@@ -0,0 +1,17 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "google/gemini-3.1-flash-lite-preview"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 0.28485
+output = 1.7091
+cache_read = 0.028485
+cache_write = 0.094951
diff --git a/providers/maxlayer/models/google/gemini-3.1-flash-lite.toml b/providers/maxlayer/models/google/gemini-3.1-flash-lite.toml
new file mode 100644
index 00000000000..5eb588cca2e
--- /dev/null
+++ b/providers/maxlayer/models/google/gemini-3.1-flash-lite.toml
@@ -0,0 +1,17 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "google/gemini-3.1-flash-lite"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 0.28485
+output = 1.7091
+cache_read = 0.028485
+cache_write = 0.094951
diff --git a/providers/maxlayer/models/google/gemini-3.1-pro-preview-customtools.toml b/providers/maxlayer/models/google/gemini-3.1-pro-preview-customtools.toml
new file mode 100644
index 00000000000..a0f8d248498
--- /dev/null
+++ b/providers/maxlayer/models/google/gemini-3.1-pro-preview-customtools.toml
@@ -0,0 +1,14 @@
+base_model = "google/gemini-3.1-pro-preview-customtools"
+
+[interleaved]
+field = "reasoning_details"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 2.2788
+output = 13.6728
+cache_read = 0.22788
+cache_write = 0.427275
diff --git a/providers/maxlayer/models/google/gemini-3.1-pro-preview.toml b/providers/maxlayer/models/google/gemini-3.1-pro-preview.toml
new file mode 100644
index 00000000000..7fc760c873c
--- /dev/null
+++ b/providers/maxlayer/models/google/gemini-3.1-pro-preview.toml
@@ -0,0 +1,14 @@
+base_model = "google/gemini-3.1-pro-preview"
+
+[interleaved]
+field = "reasoning_details"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 2.2788
+output = 13.6728
+cache_read = 0.22788
+cache_write = 0.427275
diff --git a/providers/maxlayer/models/google/gemini-3.5-flash-lite.toml b/providers/maxlayer/models/google/gemini-3.5-flash-lite.toml
new file mode 100644
index 00000000000..1de9180cc2b
--- /dev/null
+++ b/providers/maxlayer/models/google/gemini-3.5-flash-lite.toml
@@ -0,0 +1,11 @@
+base_model = "google/gemini-3.5-flash-lite"
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 0.34182
+output = 2.8485
+cache_read = 0.034182
+cache_write = 0.094951
diff --git a/providers/maxlayer/models/google/gemini-3.5-flash.toml b/providers/maxlayer/models/google/gemini-3.5-flash.toml
new file mode 100644
index 00000000000..0c2bef245ae
--- /dev/null
+++ b/providers/maxlayer/models/google/gemini-3.5-flash.toml
@@ -0,0 +1,11 @@
+base_model = "google/gemini-3.5-flash"
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 1.7091
+output = 10.2546
+cache_read = 0.17091
+cache_write = 0.094951
diff --git a/providers/maxlayer/models/google/gemini-3.6-flash.toml b/providers/maxlayer/models/google/gemini-3.6-flash.toml
new file mode 100644
index 00000000000..5f910f4c833
--- /dev/null
+++ b/providers/maxlayer/models/google/gemini-3.6-flash.toml
@@ -0,0 +1,11 @@
+base_model = "google/gemini-3.6-flash"
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 0.85455
+output = 4.27275
+cache_read = 0.085455
+cache_write = 0.047475
diff --git a/providers/maxlayer/models/google/gemini-3.7-flash.toml b/providers/maxlayer/models/google/gemini-3.7-flash.toml
new file mode 100644
index 00000000000..bc01c3defac
--- /dev/null
+++ b/providers/maxlayer/models/google/gemini-3.7-flash.toml
@@ -0,0 +1,11 @@
+base_model = "google/gemini-3.7-flash"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 0.427275
+output = 2.136375
+cache_read = 0.042729
+cache_write = 0.023738
diff --git a/providers/maxlayer/models/google/gemma-4-26b-a4b-it.toml b/providers/maxlayer/models/google/gemma-4-26b-a4b-it.toml
new file mode 100644
index 00000000000..0f780bc43df
--- /dev/null
+++ b/providers/maxlayer/models/google/gemma-4-26b-a4b-it.toml
@@ -0,0 +1,17 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "google/gemma-4-26b-a4b-it"
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.079758
+output = 0.387396
+
+[limit]
+output = 16_384
+
+[modalities]
+input = ["image", "text", "video"]
diff --git a/providers/maxlayer/models/google/gemma-4-31b-it.toml b/providers/maxlayer/models/google/gemma-4-31b-it.toml
new file mode 100644
index 00000000000..ed4325ea36c
--- /dev/null
+++ b/providers/maxlayer/models/google/gemma-4-31b-it.toml
@@ -0,0 +1,17 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "google/gemma-4-31b-it"
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.11394
+output = 0.387396
+
+[limit]
+output = 262_144
+
+[modalities]
+input = ["image", "text", "video"]
diff --git a/providers/maxlayer/models/meta-llama/llama-3.1-8b-instruct.toml b/providers/maxlayer/models/meta-llama/llama-3.1-8b-instruct.toml
new file mode 100644
index 00000000000..1f9f42a8804
--- /dev/null
+++ b/providers/maxlayer/models/meta-llama/llama-3.1-8b-instruct.toml
@@ -0,0 +1,11 @@
+base_model = "meta/llama-3.1-8b-instruct"
+structured_output = true
+
+[cost]
+input = 0.05697
+output = 0.091152
+cache_read = 0.028485
+
+[limit]
+context = 131_072
+output = 131_072
diff --git a/providers/maxlayer/models/meta-llama/llama-3.3-70b-instruct.toml b/providers/maxlayer/models/meta-llama/llama-3.3-70b-instruct.toml
new file mode 100644
index 00000000000..fc969b8ec16
--- /dev/null
+++ b/providers/maxlayer/models/meta-llama/llama-3.3-70b-instruct.toml
@@ -0,0 +1,11 @@
+base_model = "meta/llama-3.3-70b-instruct"
+attachment = false
+structured_output = true
+
+[cost]
+input = 0.11394
+output = 0.364608
+
+[limit]
+context = 131_072
+output = 16_384
diff --git a/providers/maxlayer/models/meta/muse-glimmer-30b.toml b/providers/maxlayer/models/meta/muse-glimmer-30b.toml
new file mode 100644
index 00000000000..c8b99b7f62d
--- /dev/null
+++ b/providers/maxlayer/models/meta/muse-glimmer-30b.toml
@@ -0,0 +1,10 @@
+base_model = "meta/muse-glimmer-30b"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "xhigh"]
+
+[cost]
+input = 0.34182
+output = 1.25334
+cache_read = 0.045576
diff --git a/providers/maxlayer/models/meta/muse-spark-1.1.toml b/providers/maxlayer/models/meta/muse-spark-1.1.toml
new file mode 100644
index 00000000000..b6b4cfb5934
--- /dev/null
+++ b/providers/maxlayer/models/meta/muse-spark-1.1.toml
@@ -0,0 +1,17 @@
+base_model = "meta/muse-spark-1.1"
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high", "xhigh"]
+
+[cost]
+input = 1.42425
+output = 4.84245
+cache_read = 0.17091
+
+[limit]
+context = 1_048_576
+output = 1_048_576
+
+[modalities]
+input = ["text", "image", "video", "pdf", "audio"]
diff --git a/providers/maxlayer/models/meta/muse-spark-1.2.toml b/providers/maxlayer/models/meta/muse-spark-1.2.toml
new file mode 100644
index 00000000000..10887a5220b
--- /dev/null
+++ b/providers/maxlayer/models/meta/muse-spark-1.2.toml
@@ -0,0 +1,13 @@
+base_model = "meta/muse-spark-1.2"
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high", "xhigh"]
+
+[cost]
+input = 1.42425
+output = 4.84245
+cache_read = 0.17091
+
+[limit]
+output = 1_048_576
diff --git a/providers/maxlayer/models/minimax/minimax-m2-her.toml b/providers/maxlayer/models/minimax/minimax-m2-her.toml
new file mode 100644
index 00000000000..82fd119d20e
--- /dev/null
+++ b/providers/maxlayer/models/minimax/minimax-m2-her.toml
@@ -0,0 +1,13 @@
+base_model = "minimax/MiniMax-M2-Her"
+reasoning = false
+tool_call = false
+structured_output = false
+
+[cost]
+input = 0.34182
+output = 1.36728
+cache_read = 0.034182
+
+[limit]
+context = 65_536
+output = 2_048
diff --git a/providers/maxlayer/models/minimax/minimax-m2.1.toml b/providers/maxlayer/models/minimax/minimax-m2.1.toml
new file mode 100644
index 00000000000..a6e7c5bbcad
--- /dev/null
+++ b/providers/maxlayer/models/minimax/minimax-m2.1.toml
@@ -0,0 +1,11 @@
+base_model = "minimax/MiniMax-M2.1"
+structured_output = false
+reasoning_options = []
+
+[interleaved]
+field = "reasoning_details"
+
+[cost]
+input = 0.34182
+output = 1.36728
+cache_read = 0.034182
diff --git a/providers/maxlayer/models/minimax/minimax-m2.5.toml b/providers/maxlayer/models/minimax/minimax-m2.5.toml
new file mode 100644
index 00000000000..2692b375e70
--- /dev/null
+++ b/providers/maxlayer/models/minimax/minimax-m2.5.toml
@@ -0,0 +1,14 @@
+base_model = "minimax/MiniMax-M2.5"
+structured_output = true
+reasoning_options = []
+
+[interleaved]
+field = "reasoning_details"
+
+[cost]
+input = 0.256365
+output = 1.02546
+cache_read = 0.068364
+
+[limit]
+output = 196_608
diff --git a/providers/maxlayer/models/minimax/minimax-m2.7.toml b/providers/maxlayer/models/minimax/minimax-m2.7.toml
new file mode 100644
index 00000000000..aa2baad8517
--- /dev/null
+++ b/providers/maxlayer/models/minimax/minimax-m2.7.toml
@@ -0,0 +1,8 @@
+base_model = "minimax/MiniMax-M2.7"
+structured_output = true
+reasoning_options = []
+
+[cost]
+input = 0.34182
+output = 1.36728
+cache_read = 0.068364
diff --git a/providers/maxlayer/models/minimax/minimax-m2.toml b/providers/maxlayer/models/minimax/minimax-m2.toml
new file mode 100644
index 00000000000..612e18b13d1
--- /dev/null
+++ b/providers/maxlayer/models/minimax/minimax-m2.toml
@@ -0,0 +1,14 @@
+base_model = "minimax/MiniMax-M2"
+structured_output = true
+reasoning_options = []
+
+[interleaved]
+field = "reasoning_details"
+
+[cost]
+input = 0.290547
+output = 1.162188
+
+[limit]
+context = 204_800
+output = 131_072
diff --git a/providers/maxlayer/models/minimax/minimax-m3.toml b/providers/maxlayer/models/minimax/minimax-m3.toml
new file mode 100644
index 00000000000..3d9c17e2b68
--- /dev/null
+++ b/providers/maxlayer/models/minimax/minimax-m3.toml
@@ -0,0 +1,12 @@
+base_model = "minimax/MiniMax-M3"
+structured_output = true
+reasoning_options = []
+
+[cost]
+input = 0.34182
+output = 1.36728
+cache_read = 0.068364
+
+[limit]
+context = 1_048_576
+output = 512_000
diff --git a/providers/maxlayer/models/mistralai/mistral-large-2512.toml b/providers/maxlayer/models/mistralai/mistral-large-2512.toml
new file mode 100644
index 00000000000..206a0180189
--- /dev/null
+++ b/providers/maxlayer/models/mistralai/mistral-large-2512.toml
@@ -0,0 +1,10 @@
+base_model = "mistral/mistral-large-2512"
+structured_output = true
+
+[cost]
+input = 0.5697
+output = 1.7091
+cache_read = 0.05697
+
+[modalities]
+input = ["text", "image", "pdf"]
diff --git a/providers/maxlayer/models/mistralai/mistral-nemo.toml b/providers/maxlayer/models/mistralai/mistral-nemo.toml
new file mode 100644
index 00000000000..bd7b643fc5f
--- /dev/null
+++ b/providers/maxlayer/models/mistralai/mistral-nemo.toml
@@ -0,0 +1,10 @@
+base_model = "mistral/mistral-nemo"
+structured_output = true
+
+[cost]
+input = 0.021649
+output = 0.034182
+
+[limit]
+context = 131_072
+output = 16_384
diff --git a/providers/maxlayer/models/mistralai/mistral-small-2603.toml b/providers/maxlayer/models/mistralai/mistral-small-2603.toml
new file mode 100644
index 00000000000..8a601822836
--- /dev/null
+++ b/providers/maxlayer/models/mistralai/mistral-small-2603.toml
@@ -0,0 +1,15 @@
+base_model = "mistral/mistral-small-2603"
+structured_output = true
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "high"]
+
+[cost]
+input = 0.17091
+output = 0.68364
+cache_read = 0.017091
+
+[limit]
+context = 262_144
+output = 262_144
diff --git a/providers/maxlayer/models/moonshotai/kimi-k2-thinking.toml b/providers/maxlayer/models/moonshotai/kimi-k2-thinking.toml
new file mode 100644
index 00000000000..bcf9a07e85d
--- /dev/null
+++ b/providers/maxlayer/models/moonshotai/kimi-k2-thinking.toml
@@ -0,0 +1,14 @@
+base_model = "moonshotai/kimi-k2-thinking"
+structured_output = true
+reasoning_options = []
+
+[interleaved]
+field = "reasoning_details"
+
+[cost]
+input = 0.68364
+output = 2.8485
+cache_read = 0.17091
+
+[limit]
+output = 100_352
diff --git a/providers/maxlayer/models/moonshotai/kimi-k2.5.toml b/providers/maxlayer/models/moonshotai/kimi-k2.5.toml
new file mode 100644
index 00000000000..9e44980ba45
--- /dev/null
+++ b/providers/maxlayer/models/moonshotai/kimi-k2.5.toml
@@ -0,0 +1,14 @@
+base_model = "moonshotai/kimi-k2.5"
+temperature = true
+reasoning_options = []
+
+[interleaved]
+field = "reasoning_details"
+
+[cost]
+input = 0.51273
+output = 2.56365
+cache_read = 0.079758
+
+[modalities]
+input = ["text", "image"]
diff --git a/providers/maxlayer/models/moonshotai/kimi-k2.6.toml b/providers/maxlayer/models/moonshotai/kimi-k2.6.toml
new file mode 100644
index 00000000000..4b0c9fbad0d
--- /dev/null
+++ b/providers/maxlayer/models/moonshotai/kimi-k2.6.toml
@@ -0,0 +1,13 @@
+base_model = "moonshotai/kimi-k2.6"
+reasoning_options = []
+
+[interleaved]
+field = "reasoning_details"
+
+[cost]
+input = 1.08243
+output = 4.5576
+cache_read = 0.182304
+
+[modalities]
+input = ["text", "image"]
diff --git a/providers/maxlayer/models/moonshotai/kimi-k2.7-code.toml b/providers/maxlayer/models/moonshotai/kimi-k2.7-code.toml
new file mode 100644
index 00000000000..0dbecb7f548
--- /dev/null
+++ b/providers/maxlayer/models/moonshotai/kimi-k2.7-code.toml
@@ -0,0 +1,11 @@
+base_model = "moonshotai/kimi-k2.7-code"
+temperature = true
+reasoning_options = []
+
+[cost]
+input = 0.763398
+output = 3.87396
+cache_read = 0.193698
+
+[modalities]
+input = ["text", "image"]
diff --git a/providers/maxlayer/models/moonshotai/kimi-k3.toml b/providers/maxlayer/models/moonshotai/kimi-k3.toml
new file mode 100644
index 00000000000..3e455d5dde5
--- /dev/null
+++ b/providers/maxlayer/models/moonshotai/kimi-k3.toml
@@ -0,0 +1,20 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "moonshotai/kimi-k3"
+temperature = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "high", "max"]
+
+[cost]
+input = 3.4182
+output = 17.091
+cache_read = 0.34182
+
+[limit]
+output = 1_048_576
diff --git a/providers/maxlayer/models/nvidia/nemotron-3-nano-30b-a3b.toml b/providers/maxlayer/models/nvidia/nemotron-3-nano-30b-a3b.toml
new file mode 100644
index 00000000000..8546a5804be
--- /dev/null
+++ b/providers/maxlayer/models/nvidia/nemotron-3-nano-30b-a3b.toml
@@ -0,0 +1,8 @@
+base_model = "nvidia/nemotron-3-nano-30b-a3b"
+structured_output = true
+reasoning_options = []
+
+[cost]
+input = 0.05697
+output = 0.22788
+cache_read = 0.034182
diff --git a/providers/maxlayer/models/nvidia/nemotron-3-super-120b-a12b.toml b/providers/maxlayer/models/nvidia/nemotron-3-super-120b-a12b.toml
new file mode 100644
index 00000000000..4eac3b9a2e9
--- /dev/null
+++ b/providers/maxlayer/models/nvidia/nemotron-3-super-120b-a12b.toml
@@ -0,0 +1,23 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "nvidia/nemotron-3-super-120b-a12b"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium"]
+
+[[reasoning_options]]
+type = "budget_tokens"
+
+[cost]
+input = 0.096849
+output = 0.45576
+
+[limit]
+context = 1_000_000
+output = 16_384
diff --git a/providers/maxlayer/models/nvidia/nemotron-3-ultra-550b-a55b.toml b/providers/maxlayer/models/nvidia/nemotron-3-ultra-550b-a55b.toml
new file mode 100644
index 00000000000..6c6649f74b3
--- /dev/null
+++ b/providers/maxlayer/models/nvidia/nemotron-3-ultra-550b-a55b.toml
@@ -0,0 +1,24 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "nvidia/nemotron-3-ultra-550b-a55b"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["medium", "high"]
+
+[[reasoning_options]]
+type = "budget_tokens"
+
+[cost]
+input = 0.68364
+output = 4.10184
+cache_read = 0.22788
+
+[limit]
+context = 512_288
+output = 16_384
diff --git a/providers/maxlayer/models/nvidia/nemotron-3.5-lightning.toml b/providers/maxlayer/models/nvidia/nemotron-3.5-lightning.toml
new file mode 100644
index 00000000000..ca36fd12635
--- /dev/null
+++ b/providers/maxlayer/models/nvidia/nemotron-3.5-lightning.toml
@@ -0,0 +1,15 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "nvidia/nemotron-3.5-lightning"
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.091152
+output = 0.22788
+cache_read = 0.045576
+
+[limit]
+output = 131_072
diff --git a/providers/maxlayer/models/openai/gpt-3.5-turbo.toml b/providers/maxlayer/models/openai/gpt-3.5-turbo.toml
new file mode 100644
index 00000000000..d1bfe66eb12
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-3.5-turbo.toml
@@ -0,0 +1,7 @@
+base_model = "openai/gpt-3.5-turbo"
+tool_call = true
+structured_output = true
+
+[cost]
+input = 0.5697
+output = 1.7091
diff --git a/providers/maxlayer/models/openai/gpt-4-turbo.toml b/providers/maxlayer/models/openai/gpt-4-turbo.toml
new file mode 100644
index 00000000000..c2307bdc743
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-4-turbo.toml
@@ -0,0 +1,6 @@
+base_model = "openai/gpt-4-turbo"
+structured_output = true
+
+[cost]
+input = 11.394
+output = 34.182
diff --git a/providers/maxlayer/models/openai/gpt-4.1-mini.toml b/providers/maxlayer/models/openai/gpt-4.1-mini.toml
new file mode 100644
index 00000000000..7f2c03771ed
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-4.1-mini.toml
@@ -0,0 +1,6 @@
+base_model = "openai/gpt-4.1-mini"
+
+[cost]
+input = 0.45576
+output = 1.82304
+cache_read = 0.11394
diff --git a/providers/maxlayer/models/openai/gpt-4.1-nano.toml b/providers/maxlayer/models/openai/gpt-4.1-nano.toml
new file mode 100644
index 00000000000..c702fe4b6ed
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-4.1-nano.toml
@@ -0,0 +1,9 @@
+base_model = "openai/gpt-4.1-nano"
+
+[cost]
+input = 0.11394
+output = 0.45576
+cache_read = 0.028485
+
+[modalities]
+input = ["image", "text", "pdf"]
diff --git a/providers/maxlayer/models/openai/gpt-4.1.toml b/providers/maxlayer/models/openai/gpt-4.1.toml
new file mode 100644
index 00000000000..ed3fb4ae318
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-4.1.toml
@@ -0,0 +1,6 @@
+base_model = "openai/gpt-4.1"
+
+[cost]
+input = 2.2788
+output = 9.1152
+cache_read = 0.5697
diff --git a/providers/maxlayer/models/openai/gpt-4.toml b/providers/maxlayer/models/openai/gpt-4.toml
new file mode 100644
index 00000000000..9526da8b21c
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-4.toml
@@ -0,0 +1,11 @@
+base_model = "openai/gpt-4"
+attachment = false
+structured_output = true
+
+[cost]
+input = 34.182
+output = 68.364
+
+[limit]
+context = 8_191
+output = 4_096
diff --git a/providers/maxlayer/models/openai/gpt-4o-2024-05-13.toml b/providers/maxlayer/models/openai/gpt-4o-2024-05-13.toml
new file mode 100644
index 00000000000..5be38122608
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-4o-2024-05-13.toml
@@ -0,0 +1,8 @@
+base_model = "openai/gpt-4o-2024-05-13"
+
+[cost]
+input = 5.697
+output = 17.091
+
+[modalities]
+input = ["text", "image", "pdf"]
diff --git a/providers/maxlayer/models/openai/gpt-4o-2024-08-06.toml b/providers/maxlayer/models/openai/gpt-4o-2024-08-06.toml
new file mode 100644
index 00000000000..387b4df5296
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-4o-2024-08-06.toml
@@ -0,0 +1,9 @@
+base_model = "openai/gpt-4o-2024-08-06"
+
+[cost]
+input = 2.8485
+output = 11.394
+cache_read = 1.42425
+
+[modalities]
+input = ["text", "image", "pdf"]
diff --git a/providers/maxlayer/models/openai/gpt-4o-2024-11-20.toml b/providers/maxlayer/models/openai/gpt-4o-2024-11-20.toml
new file mode 100644
index 00000000000..f38235558a6
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-4o-2024-11-20.toml
@@ -0,0 +1,9 @@
+base_model = "openai/gpt-4o-2024-11-20"
+
+[cost]
+input = 2.8485
+output = 11.394
+cache_read = 1.42425
+
+[modalities]
+input = ["text", "image", "pdf"]
diff --git a/providers/maxlayer/models/openai/gpt-4o-mini.toml b/providers/maxlayer/models/openai/gpt-4o-mini.toml
new file mode 100644
index 00000000000..48251db28a3
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-4o-mini.toml
@@ -0,0 +1,6 @@
+base_model = "openai/gpt-4o-mini"
+
+[cost]
+input = 0.17091
+output = 0.68364
+cache_read = 0.085455
diff --git a/providers/maxlayer/models/openai/gpt-4o.toml b/providers/maxlayer/models/openai/gpt-4o.toml
new file mode 100644
index 00000000000..74f5c7d301a
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-4o.toml
@@ -0,0 +1,6 @@
+base_model = "openai/gpt-4o"
+
+[cost]
+input = 2.8485
+output = 11.394
+cache_read = 1.42425
diff --git a/providers/maxlayer/models/openai/gpt-5-mini.toml b/providers/maxlayer/models/openai/gpt-5-mini.toml
new file mode 100644
index 00000000000..b11b3ff627c
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5-mini.toml
@@ -0,0 +1,13 @@
+base_model = "openai/gpt-5-mini"
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 0.28485
+output = 2.2788
+cache_read = 0.028485
+
+[modalities]
+input = ["text", "image", "pdf"]
diff --git a/providers/maxlayer/models/openai/gpt-5-nano.toml b/providers/maxlayer/models/openai/gpt-5-nano.toml
new file mode 100644
index 00000000000..7ae06057bb8
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5-nano.toml
@@ -0,0 +1,13 @@
+base_model = "openai/gpt-5-nano"
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 0.05697
+output = 0.45576
+cache_read = 0.005697
+
+[modalities]
+input = ["text", "image", "pdf"]
diff --git a/providers/maxlayer/models/openai/gpt-5-pro.toml b/providers/maxlayer/models/openai/gpt-5-pro.toml
new file mode 100644
index 00000000000..cf9979854e1
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5-pro.toml
@@ -0,0 +1,15 @@
+base_model = "openai/gpt-5-pro"
+
+[[reasoning_options]]
+type = "effort"
+values = ["high"]
+
+[cost]
+input = 17.091
+output = 136.728
+
+[limit]
+output = 128_000
+
+[modalities]
+input = ["image", "text", "pdf"]
diff --git a/providers/maxlayer/models/openai/gpt-5.1-codex-max.toml b/providers/maxlayer/models/openai/gpt-5.1-codex-max.toml
new file mode 100644
index 00000000000..ba54dbe9c61
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.1-codex-max.toml
@@ -0,0 +1,10 @@
+base_model = "openai/gpt-5.1-codex-max"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "xhigh"]
+
+[cost]
+input = 1.42425
+output = 11.394
+cache_read = 0.142425
diff --git a/providers/maxlayer/models/openai/gpt-5.1-codex-mini.toml b/providers/maxlayer/models/openai/gpt-5.1-codex-mini.toml
new file mode 100644
index 00000000000..f516e6d9495
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.1-codex-mini.toml
@@ -0,0 +1,16 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "openai/gpt-5.1-codex-mini"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 0.28485
+output = 2.2788
+cache_read = 0.034182
diff --git a/providers/maxlayer/models/openai/gpt-5.1-codex.toml b/providers/maxlayer/models/openai/gpt-5.1-codex.toml
new file mode 100644
index 00000000000..0be994cb7ec
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.1-codex.toml
@@ -0,0 +1,10 @@
+base_model = "openai/gpt-5.1-codex"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 1.42425
+output = 11.394
+cache_read = 0.148122
diff --git a/providers/maxlayer/models/openai/gpt-5.1.toml b/providers/maxlayer/models/openai/gpt-5.1.toml
new file mode 100644
index 00000000000..81864feafdd
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.1.toml
@@ -0,0 +1,13 @@
+base_model = "openai/gpt-5.1"
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "medium", "high"]
+
+[cost]
+input = 1.42425
+output = 11.394
+cache_read = 0.142425
+
+[modalities]
+input = ["image", "text", "pdf"]
diff --git a/providers/maxlayer/models/openai/gpt-5.2-codex.toml b/providers/maxlayer/models/openai/gpt-5.2-codex.toml
new file mode 100644
index 00000000000..e5621c60e12
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.2-codex.toml
@@ -0,0 +1,13 @@
+base_model = "openai/gpt-5.2-codex"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "xhigh"]
+
+[cost]
+input = 1.99395
+output = 15.9516
+cache_read = 0.199395
+
+[modalities]
+input = ["text", "image"]
diff --git a/providers/maxlayer/models/openai/gpt-5.2-pro.toml b/providers/maxlayer/models/openai/gpt-5.2-pro.toml
new file mode 100644
index 00000000000..eb10ebeb052
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.2-pro.toml
@@ -0,0 +1,13 @@
+base_model = "openai/gpt-5.2-pro"
+structured_output = true
+
+[[reasoning_options]]
+type = "effort"
+values = ["medium", "high", "xhigh"]
+
+[cost]
+input = 23.9274
+output = 191.4192
+
+[modalities]
+input = ["image", "text", "pdf"]
diff --git a/providers/maxlayer/models/openai/gpt-5.2.toml b/providers/maxlayer/models/openai/gpt-5.2.toml
new file mode 100644
index 00000000000..689dfaedbcd
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.2.toml
@@ -0,0 +1,13 @@
+base_model = "openai/gpt-5.2"
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "medium", "high", "xhigh"]
+
+[cost]
+input = 1.99395
+output = 15.9516
+cache_read = 0.199395
+
+[modalities]
+input = ["pdf", "image", "text"]
diff --git a/providers/maxlayer/models/openai/gpt-5.3-codex.toml b/providers/maxlayer/models/openai/gpt-5.3-codex.toml
new file mode 100644
index 00000000000..d2a5d8f3842
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.3-codex.toml
@@ -0,0 +1,10 @@
+base_model = "openai/gpt-5.3-codex"
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "medium", "high", "xhigh"]
+
+[cost]
+input = 1.99395
+output = 15.9516
+cache_read = 0.199395
diff --git a/providers/maxlayer/models/openai/gpt-5.4-mini.toml b/providers/maxlayer/models/openai/gpt-5.4-mini.toml
new file mode 100644
index 00000000000..02dc10b860b
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.4-mini.toml
@@ -0,0 +1,13 @@
+base_model = "openai/gpt-5.4-mini"
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "medium", "high", "xhigh"]
+
+[cost]
+input = 0.85455
+output = 5.1273
+cache_read = 0.085455
+
+[modalities]
+input = ["pdf", "image", "text"]
diff --git a/providers/maxlayer/models/openai/gpt-5.4-nano.toml b/providers/maxlayer/models/openai/gpt-5.4-nano.toml
new file mode 100644
index 00000000000..ef6e56e8afd
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.4-nano.toml
@@ -0,0 +1,13 @@
+base_model = "openai/gpt-5.4-nano"
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "medium", "high", "xhigh"]
+
+[cost]
+input = 0.22788
+output = 1.42425
+cache_read = 0.022788
+
+[modalities]
+input = ["pdf", "image", "text"]
diff --git a/providers/maxlayer/models/openai/gpt-5.4-pro.toml b/providers/maxlayer/models/openai/gpt-5.4-pro.toml
new file mode 100644
index 00000000000..295687af04d
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.4-pro.toml
@@ -0,0 +1,13 @@
+base_model = "openai/gpt-5.4-pro"
+structured_output = true
+
+[[reasoning_options]]
+type = "effort"
+values = ["medium", "high", "xhigh"]
+
+[cost]
+input = 34.182
+output = 205.092
+
+[modalities]
+input = ["text", "image", "pdf"]
diff --git a/providers/maxlayer/models/openai/gpt-5.4.toml b/providers/maxlayer/models/openai/gpt-5.4.toml
new file mode 100644
index 00000000000..98040b08164
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.4.toml
@@ -0,0 +1,10 @@
+base_model = "openai/gpt-5.4"
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "medium", "high", "xhigh"]
+
+[cost]
+input = 2.8485
+output = 17.091
+cache_read = 0.28485
diff --git a/providers/maxlayer/models/openai/gpt-5.5-pro.toml b/providers/maxlayer/models/openai/gpt-5.5-pro.toml
new file mode 100644
index 00000000000..13047ac295c
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.5-pro.toml
@@ -0,0 +1,9 @@
+base_model = "openai/gpt-5.5-pro"
+
+[[reasoning_options]]
+type = "effort"
+values = ["medium", "high", "xhigh"]
+
+[cost]
+input = 34.182
+output = 205.092
diff --git a/providers/maxlayer/models/openai/gpt-5.5.toml b/providers/maxlayer/models/openai/gpt-5.5.toml
new file mode 100644
index 00000000000..4f1ab64a264
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.5.toml
@@ -0,0 +1,10 @@
+base_model = "openai/gpt-5.5"
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "medium", "high", "xhigh"]
+
+[cost]
+input = 5.697
+output = 34.182
+cache_read = 0.5697
diff --git a/providers/maxlayer/models/openai/gpt-5.6-luna-pro.toml b/providers/maxlayer/models/openai/gpt-5.6-luna-pro.toml
new file mode 100644
index 00000000000..33f89407d99
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.6-luna-pro.toml
@@ -0,0 +1,12 @@
+base_model = "openai/gpt-5.6-luna"
+name = "GPT-5.6 Luna Pro"
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 0.22788
+output = 1.36728
+cache_read = 0.022788
+cache_write = 0.28485
diff --git a/providers/maxlayer/models/openai/gpt-5.6-luna.toml b/providers/maxlayer/models/openai/gpt-5.6-luna.toml
new file mode 100644
index 00000000000..72724392551
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.6-luna.toml
@@ -0,0 +1,11 @@
+base_model = "openai/gpt-5.6-luna"
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 0.22788
+output = 1.36728
+cache_read = 0.022788
+cache_write = 0.28485
diff --git a/providers/maxlayer/models/openai/gpt-5.6-sol-pro.toml b/providers/maxlayer/models/openai/gpt-5.6-sol-pro.toml
new file mode 100644
index 00000000000..7099a9c09fd
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.6-sol-pro.toml
@@ -0,0 +1,12 @@
+base_model = "openai/gpt-5.6-sol"
+name = "GPT-5.6 Sol Pro"
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 2.8485
+output = 17.091
+cache_read = 0.28485
+cache_write = 3.560625
diff --git a/providers/maxlayer/models/openai/gpt-5.6-sol.toml b/providers/maxlayer/models/openai/gpt-5.6-sol.toml
new file mode 100644
index 00000000000..a2e4b448d62
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.6-sol.toml
@@ -0,0 +1,11 @@
+base_model = "openai/gpt-5.6-sol"
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 2.8485
+output = 17.091
+cache_read = 0.28485
+cache_write = 3.560625
diff --git a/providers/maxlayer/models/openai/gpt-5.6-terra-pro.toml b/providers/maxlayer/models/openai/gpt-5.6-terra-pro.toml
new file mode 100644
index 00000000000..2cb4c3926ad
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.6-terra-pro.toml
@@ -0,0 +1,12 @@
+base_model = "openai/gpt-5.6-terra"
+name = "GPT-5.6 Terra Pro"
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 2.2788
+output = 13.6728
+cache_read = 0.22788
+cache_write = 2.8485
diff --git a/providers/maxlayer/models/openai/gpt-5.6-terra.toml b/providers/maxlayer/models/openai/gpt-5.6-terra.toml
new file mode 100644
index 00000000000..d5d31289e7e
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.6-terra.toml
@@ -0,0 +1,11 @@
+base_model = "openai/gpt-5.6-terra"
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 2.2788
+output = 13.6728
+cache_read = 0.22788
+cache_write = 2.8485
diff --git a/providers/maxlayer/models/openai/gpt-5.toml b/providers/maxlayer/models/openai/gpt-5.toml
new file mode 100644
index 00000000000..1b03e11c1e1
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-5.toml
@@ -0,0 +1,13 @@
+base_model = "openai/gpt-5"
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 1.42425
+output = 11.394
+cache_read = 0.142425
+
+[modalities]
+input = ["text", "image", "pdf"]
diff --git a/providers/maxlayer/models/openai/gpt-oss-120b.toml b/providers/maxlayer/models/openai/gpt-oss-120b.toml
new file mode 100644
index 00000000000..b40fbc8b6ac
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-oss-120b.toml
@@ -0,0 +1,12 @@
+base_model = "openai/gpt-oss-120b"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 0.034182
+output = 0.193698
+
+[limit]
+output = 131_072
diff --git a/providers/maxlayer/models/openai/gpt-oss-20b.toml b/providers/maxlayer/models/openai/gpt-oss-20b.toml
new file mode 100644
index 00000000000..f36eb1a6478
--- /dev/null
+++ b/providers/maxlayer/models/openai/gpt-oss-20b.toml
@@ -0,0 +1,12 @@
+base_model = "openai/gpt-oss-20b"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 0.034182
+output = 0.148122
+
+[limit]
+output = 131_072
diff --git a/providers/maxlayer/models/openai/o1-pro.toml b/providers/maxlayer/models/openai/o1-pro.toml
new file mode 100644
index 00000000000..6d14a8c097b
--- /dev/null
+++ b/providers/maxlayer/models/openai/o1-pro.toml
@@ -0,0 +1,13 @@
+base_model = "openai/o1-pro"
+tool_call = false
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 170.91
+output = 683.64
+
+[modalities]
+input = ["text", "image", "pdf"]
diff --git a/providers/maxlayer/models/openai/o1.toml b/providers/maxlayer/models/openai/o1.toml
new file mode 100644
index 00000000000..5740d2742e2
--- /dev/null
+++ b/providers/maxlayer/models/openai/o1.toml
@@ -0,0 +1,10 @@
+base_model = "openai/o1"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 17.091
+output = 68.364
+cache_read = 8.5455
diff --git a/providers/maxlayer/models/openai/o3-mini.toml b/providers/maxlayer/models/openai/o3-mini.toml
new file mode 100644
index 00000000000..f17f8bbcacd
--- /dev/null
+++ b/providers/maxlayer/models/openai/o3-mini.toml
@@ -0,0 +1,14 @@
+base_model = "openai/o3-mini"
+attachment = true
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 1.25334
+output = 5.01336
+cache_read = 0.62667
+
+[modalities]
+input = ["text", "pdf"]
diff --git a/providers/maxlayer/models/openai/o3-pro.toml b/providers/maxlayer/models/openai/o3-pro.toml
new file mode 100644
index 00000000000..b2752785ec8
--- /dev/null
+++ b/providers/maxlayer/models/openai/o3-pro.toml
@@ -0,0 +1,12 @@
+base_model = "openai/o3-pro"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 22.788
+output = 91.152
+
+[modalities]
+input = ["text", "pdf", "image"]
diff --git a/providers/maxlayer/models/openai/o3.toml b/providers/maxlayer/models/openai/o3.toml
new file mode 100644
index 00000000000..df4188ba153
--- /dev/null
+++ b/providers/maxlayer/models/openai/o3.toml
@@ -0,0 +1,10 @@
+base_model = "openai/o3"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 2.2788
+output = 9.1152
+cache_read = 0.5697
diff --git a/providers/maxlayer/models/openai/o4-mini.toml b/providers/maxlayer/models/openai/o4-mini.toml
new file mode 100644
index 00000000000..c5264c71c88
--- /dev/null
+++ b/providers/maxlayer/models/openai/o4-mini.toml
@@ -0,0 +1,13 @@
+base_model = "openai/o4-mini"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 1.25334
+output = 5.01336
+cache_read = 0.313335
+
+[modalities]
+input = ["image", "text", "pdf"]
diff --git a/providers/maxlayer/models/qwen/qwen-plus.toml b/providers/maxlayer/models/qwen/qwen-plus.toml
new file mode 100644
index 00000000000..09e438dcdbb
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen-plus.toml
@@ -0,0 +1,9 @@
+base_model = "alibaba/qwen-plus"
+reasoning = false
+structured_output = true
+
+[cost]
+input = 0.296244
+output = 0.888732
+cache_read = 0.059249
+cache_write = 0.370305
diff --git a/providers/maxlayer/models/qwen/qwen3-235b-a22b.toml b/providers/maxlayer/models/qwen/qwen3-235b-a22b.toml
new file mode 100644
index 00000000000..7ce74beb4c2
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3-235b-a22b.toml
@@ -0,0 +1,20 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3-235b-a22b"
+structured_output = false
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 1
+max = 38_912
+
+[cost]
+input = 0.518427
+output = 2.073708
+
+[limit]
+output = 8_192
diff --git a/providers/maxlayer/models/qwen/qwen3-30b-a3b.toml b/providers/maxlayer/models/qwen/qwen3-30b-a3b.toml
new file mode 100644
index 00000000000..e1921aa5d78
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3-30b-a3b.toml
@@ -0,0 +1,15 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3-30b-a3b"
+structured_output = false
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.148122
+output = 0.592488
+
+[limit]
+output = 8_192
diff --git a/providers/maxlayer/models/qwen/qwen3-32b.toml b/providers/maxlayer/models/qwen/qwen3-32b.toml
new file mode 100644
index 00000000000..7073995eaf6
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3-32b.toml
@@ -0,0 +1,12 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3-32b"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.091152
+output = 0.319032
diff --git a/providers/maxlayer/models/qwen/qwen3-coder-30b-a3b-instruct.toml b/providers/maxlayer/models/qwen/qwen3-coder-30b-a3b-instruct.toml
new file mode 100644
index 00000000000..4a4e2504e1f
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3-coder-30b-a3b-instruct.toml
@@ -0,0 +1,9 @@
+base_model = "alibaba/qwen3-coder-30b-a3b-instruct"
+structured_output = true
+
+[cost]
+input = 0.079758
+output = 0.319032
+
+[limit]
+output = 262_144
diff --git a/providers/maxlayer/models/qwen/qwen3-coder-flash.toml b/providers/maxlayer/models/qwen/qwen3-coder-flash.toml
new file mode 100644
index 00000000000..a04be0c4d53
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3-coder-flash.toml
@@ -0,0 +1,8 @@
+base_model = "alibaba/qwen3-coder-flash"
+structured_output = false
+
+[cost]
+input = 0.222183
+output = 1.110915
+cache_read = 0.044437
+cache_write = 0.27773
diff --git a/providers/maxlayer/models/qwen/qwen3-coder-next.toml b/providers/maxlayer/models/qwen/qwen3-coder-next.toml
new file mode 100644
index 00000000000..477b449b2c3
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3-coder-next.toml
@@ -0,0 +1,9 @@
+base_model = "alibaba/qwen3-coder-next"
+
+[cost]
+input = 0.136728
+output = 0.91152
+cache_read = 0.079758
+
+[limit]
+output = 262_144
diff --git a/providers/maxlayer/models/qwen/qwen3-coder-plus.toml b/providers/maxlayer/models/qwen/qwen3-coder-plus.toml
new file mode 100644
index 00000000000..6bf91745af2
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3-coder-plus.toml
@@ -0,0 +1,11 @@
+base_model = "alibaba/qwen3-coder-plus"
+structured_output = true
+
+[cost]
+input = 0.74061
+output = 3.70305
+cache_read = 0.148122
+cache_write = 0.925764
+
+[limit]
+context = 1_000_000
diff --git a/providers/maxlayer/models/qwen/qwen3-max.toml b/providers/maxlayer/models/qwen/qwen3-max.toml
new file mode 100644
index 00000000000..879b5b7fac5
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3-max.toml
@@ -0,0 +1,8 @@
+base_model = "alibaba/qwen3-max"
+structured_output = true
+
+[cost]
+input = 0.888732
+output = 4.44366
+cache_read = 0.177747
+cache_write = 1.110915
diff --git a/providers/maxlayer/models/qwen/qwen3-next-80b-a3b-instruct.toml b/providers/maxlayer/models/qwen/qwen3-next-80b-a3b-instruct.toml
new file mode 100644
index 00000000000..f2b54dce696
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3-next-80b-a3b-instruct.toml
@@ -0,0 +1,10 @@
+base_model = "alibaba/qwen3-next-80b-a3b-instruct"
+structured_output = true
+
+[cost]
+input = 0.102546
+output = 1.25334
+
+[limit]
+context = 262_144
+output = 16_384
diff --git a/providers/maxlayer/models/qwen/qwen3-next-80b-a3b-thinking.toml b/providers/maxlayer/models/qwen/qwen3-next-80b-a3b-thinking.toml
new file mode 100644
index 00000000000..f7ba2e692e6
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3-next-80b-a3b-thinking.toml
@@ -0,0 +1,10 @@
+base_model = "alibaba/qwen3-next-80b-a3b-thinking"
+structured_output = true
+reasoning_options = []
+
+[cost]
+input = 0.17091
+output = 1.36728
+
+[limit]
+context = 262_144
diff --git a/providers/maxlayer/models/qwen/qwen3-vl-235b-a22b-instruct.toml b/providers/maxlayer/models/qwen/qwen3-vl-235b-a22b-instruct.toml
new file mode 100644
index 00000000000..d1cdce17616
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3-vl-235b-a22b-instruct.toml
@@ -0,0 +1,9 @@
+base_model = "alibaba/qwen3-vl-235b-a22b-instruct"
+
+[cost]
+input = 0.239274
+output = 2.16486
+cache_read = 0.11394
+
+[limit]
+context = 262_144
diff --git a/providers/maxlayer/models/qwen/qwen3-vl-235b-a22b-thinking.toml b/providers/maxlayer/models/qwen/qwen3-vl-235b-a22b-thinking.toml
new file mode 100644
index 00000000000..394951ec38f
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3-vl-235b-a22b-thinking.toml
@@ -0,0 +1,6 @@
+base_model = "alibaba/qwen3-vl-235b-a22b-thinking"
+reasoning_options = []
+
+[cost]
+input = 0.45576
+output = 4.5576
diff --git a/providers/maxlayer/models/qwen/qwen3.5-122b-a10b.toml b/providers/maxlayer/models/qwen/qwen3.5-122b-a10b.toml
new file mode 100644
index 00000000000..73f139f0aa7
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.5-122b-a10b.toml
@@ -0,0 +1,14 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3.5-122b-a10b"
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.296244
+output = 2.369952
+
+[modalities]
+input = ["text", "image", "video"]
diff --git a/providers/maxlayer/models/qwen/qwen3.5-27b.toml b/providers/maxlayer/models/qwen/qwen3.5-27b.toml
new file mode 100644
index 00000000000..d670016f730
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.5-27b.toml
@@ -0,0 +1,14 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3.5-27b"
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.222183
+output = 1.777464
+
+[modalities]
+input = ["text", "image", "video"]
diff --git a/providers/maxlayer/models/qwen/qwen3.5-35b-a3b.toml b/providers/maxlayer/models/qwen/qwen3.5-35b-a3b.toml
new file mode 100644
index 00000000000..c80ee624657
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.5-35b-a3b.toml
@@ -0,0 +1,17 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3.5-35b-a3b"
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.28485
+output = 1.42425
+
+[limit]
+output = 262_144
+
+[modalities]
+input = ["text", "image", "video"]
diff --git a/providers/maxlayer/models/qwen/qwen3.5-397b-a17b.toml b/providers/maxlayer/models/qwen/qwen3.5-397b-a17b.toml
new file mode 100644
index 00000000000..931e50124c0
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.5-397b-a17b.toml
@@ -0,0 +1,14 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3.5-397b-a17b"
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.444366
+output = 2.666196
+
+[modalities]
+input = ["text", "image", "video"]
diff --git a/providers/maxlayer/models/qwen/qwen3.5-9b.toml b/providers/maxlayer/models/qwen/qwen3.5-9b.toml
new file mode 100644
index 00000000000..1ed2260d0b1
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.5-9b.toml
@@ -0,0 +1,14 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3.5-9b"
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.11394
+output = 0.17091
+
+[limit]
+output = 262_144
diff --git a/providers/maxlayer/models/qwen/qwen3.6-27b.toml b/providers/maxlayer/models/qwen/qwen3.6-27b.toml
new file mode 100644
index 00000000000..50918347446
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.6-27b.toml
@@ -0,0 +1,18 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3.6-27b"
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.68364
+output = 4.10184
+cache_read = 0.136728
+
+[limit]
+output = 262_144
+
+[modalities]
+input = ["text", "image", "video"]
diff --git a/providers/maxlayer/models/qwen/qwen3.6-35b-a3b.toml b/providers/maxlayer/models/qwen/qwen3.6-35b-a3b.toml
new file mode 100644
index 00000000000..b9ee5aaf140
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.6-35b-a3b.toml
@@ -0,0 +1,18 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3.6-35b-a3b"
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.159516
+output = 1.1394
+cache_read = 0.05697
+
+[limit]
+output = 262_144
+
+[modalities]
+input = ["text", "image", "video"]
diff --git a/providers/maxlayer/models/qwen/qwen3.6-flash.toml b/providers/maxlayer/models/qwen/qwen3.6-flash.toml
new file mode 100644
index 00000000000..e902f0e336e
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.6-flash.toml
@@ -0,0 +1,17 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3.6-flash"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 1
+max = 81_920
+
+[cost]
+input = 0.213639
+output = 1.281825
+cache_write = 0.267048
diff --git a/providers/maxlayer/models/qwen/qwen3.6-max-preview.toml b/providers/maxlayer/models/qwen/qwen3.6-max-preview.toml
new file mode 100644
index 00000000000..82bccfa4c8a
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.6-max-preview.toml
@@ -0,0 +1,18 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3.6-max-preview"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 1
+max = 131_072
+
+[cost]
+input = 1.170164
+output = 7.020983
+cache_write = 1.462706
diff --git a/providers/maxlayer/models/qwen/qwen3.6-plus.toml b/providers/maxlayer/models/qwen/qwen3.6-plus.toml
new file mode 100644
index 00000000000..c30b478c887
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.6-plus.toml
@@ -0,0 +1,18 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3.6-plus"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 1
+max = 81_920
+
+[cost]
+input = 0.370305
+output = 2.22183
+cache_write = 0.462882
diff --git a/providers/maxlayer/models/qwen/qwen3.7-flash.toml b/providers/maxlayer/models/qwen/qwen3.7-flash.toml
new file mode 100644
index 00000000000..2501104b842
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.7-flash.toml
@@ -0,0 +1,11 @@
+base_model = "alibaba/qwen3.7-flash"
+structured_output = false
+
+[[reasoning_options]]
+type = "budget_tokens"
+
+[cost]
+input = 0.034182
+output = 0.148122
+cache_read = 0.006837
+cache_write = 0.043298
diff --git a/providers/maxlayer/models/qwen/qwen3.7-max.toml b/providers/maxlayer/models/qwen/qwen3.7-max.toml
new file mode 100644
index 00000000000..0f9c6c10b68
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.7-max.toml
@@ -0,0 +1,22 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3.7-max"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 1
+max = 262_144
+
+[cost]
+input = 1.680615
+output = 5.041845
+cache_read = 0.336123
+cache_write = 2.10077
+
+[limit]
+output = 131_072
diff --git a/providers/maxlayer/models/qwen/qwen3.7-plus.toml b/providers/maxlayer/models/qwen/qwen3.7-plus.toml
new file mode 100644
index 00000000000..aacc74f140d
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.7-plus.toml
@@ -0,0 +1,25 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3.7-plus"
+structured_output = true
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 1
+max = 262_144
+
+[cost]
+input = 0.364608
+output = 1.458432
+cache_read = 0.072922
+cache_write = 0.45576
+
+[limit]
+output = 131_072
+
+[modalities]
+input = ["text", "image"]
diff --git a/providers/maxlayer/models/qwen/qwen3.8-2.4t-a95b.toml b/providers/maxlayer/models/qwen/qwen3.8-2.4t-a95b.toml
new file mode 100644
index 00000000000..969134f2324
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.8-2.4t-a95b.toml
@@ -0,0 +1,14 @@
+base_model = "alibaba/qwen3.8-2.4t-a95b"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "xhigh"]
+
+[cost]
+input = 2.2788
+output = 6.8364
+cache_read = 0.28485
+
+[limit]
+context = 1_048_576
+output = 262_144
diff --git a/providers/maxlayer/models/qwen/qwen3.8-27b.toml b/providers/maxlayer/models/qwen/qwen3.8-27b.toml
new file mode 100644
index 00000000000..ac32ca6974e
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.8-27b.toml
@@ -0,0 +1,20 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "alibaba/qwen3.8-27b"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "xhigh"]
+
+[cost]
+input = 0.51273
+output = 3.64608
+cache_read = 0.05697
+
+[limit]
+context = 1_000_000
+output = 131_072
diff --git a/providers/maxlayer/models/qwen/qwen3.8-max.toml b/providers/maxlayer/models/qwen/qwen3.8-max.toml
new file mode 100644
index 00000000000..1b78bb65edd
--- /dev/null
+++ b/providers/maxlayer/models/qwen/qwen3.8-max.toml
@@ -0,0 +1,15 @@
+base_model = "alibaba/qwen3.8-max"
+structured_output = true
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high", "xhigh"]
+
+[cost]
+input = 2.2788
+output = 6.8364
+cache_read = 0.28485
+cache_write = 2.8485
+
+[modalities]
+input = ["text", "image", "video"]
diff --git a/providers/maxlayer/models/sakana/fugu-ultra.toml b/providers/maxlayer/models/sakana/fugu-ultra.toml
new file mode 100644
index 00000000000..0479f7e3443
--- /dev/null
+++ b/providers/maxlayer/models/sakana/fugu-ultra.toml
@@ -0,0 +1,13 @@
+base_model = "sakana/fugu-ultra"
+
+[[reasoning_options]]
+type = "effort"
+values = ["high", "xhigh", "max"]
+
+[cost]
+input = 5.697
+output = 34.182
+cache_read = 0.5697
+
+[limit]
+output = 128_000
diff --git a/providers/maxlayer/models/sakana/sakana-namazu.toml b/providers/maxlayer/models/sakana/sakana-namazu.toml
new file mode 100644
index 00000000000..42867a7cb81
--- /dev/null
+++ b/providers/maxlayer/models/sakana/sakana-namazu.toml
@@ -0,0 +1,11 @@
+base_model = "sakana/sakana-namazu"
+temperature = false
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "high"]
+
+[cost]
+input = 1.08243
+output = 4.5576
+cache_read = 0.17091
diff --git a/providers/maxlayer/models/stepfun/step-3.5-flash.toml b/providers/maxlayer/models/stepfun/step-3.5-flash.toml
new file mode 100644
index 00000000000..86934b4b72d
--- /dev/null
+++ b/providers/maxlayer/models/stepfun/step-3.5-flash.toml
@@ -0,0 +1,12 @@
+base_model = "stepfun/step-3.5-flash"
+base_model_omit = ["limit.input"]
+structured_output = false
+reasoning_options = []
+
+[cost]
+input = 0.11394
+output = 0.34182
+
+[limit]
+context = 262_144
+output = 65_536
diff --git a/providers/maxlayer/models/stepfun/step-3.7-flash.toml b/providers/maxlayer/models/stepfun/step-3.7-flash.toml
new file mode 100644
index 00000000000..fea56a9fb9f
--- /dev/null
+++ b/providers/maxlayer/models/stepfun/step-3.7-flash.toml
@@ -0,0 +1,15 @@
+base_model = "stepfun/step-3.7-flash"
+base_model_omit = ["limit.input"]
+structured_output = true
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 0.22788
+output = 1.31031
+cache_read = 0.045576
+
+[limit]
+context = 262_144
diff --git a/providers/maxlayer/models/tencent/hy3-preview.toml b/providers/maxlayer/models/tencent/hy3-preview.toml
new file mode 100644
index 00000000000..ec6edf401d7
--- /dev/null
+++ b/providers/maxlayer/models/tencent/hy3-preview.toml
@@ -0,0 +1,15 @@
+base_model = "tencent/hy3-preview"
+structured_output = false
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "high"]
+
+[cost]
+input = 0.205092
+output = 0.68364
+cache_read = 0.068364
+
+[limit]
+context = 262_144
+output = 262_144
diff --git a/providers/maxlayer/models/tencent/hy3.toml b/providers/maxlayer/models/tencent/hy3.toml
new file mode 100644
index 00000000000..72888ccb6de
--- /dev/null
+++ b/providers/maxlayer/models/tencent/hy3.toml
@@ -0,0 +1,15 @@
+base_model = "tencent/hy3"
+structured_output = true
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "high"]
+
+[cost]
+input = 0.150401
+output = 0.601604
+cache_read = 0.037601
+
+[limit]
+context = 262_144
+output = 128_000
diff --git a/providers/maxlayer/models/thinkingmachines/inkling-small.toml b/providers/maxlayer/models/thinkingmachines/inkling-small.toml
new file mode 100644
index 00000000000..781da77b0a9
--- /dev/null
+++ b/providers/maxlayer/models/thinkingmachines/inkling-small.toml
@@ -0,0 +1,14 @@
+base_model = "thinkingmachines/inkling-small"
+structured_output = true
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "minimal", "low", "medium", "high", "max"]
+
+[cost]
+input = 0.51273
+output = 1.36728
+cache_read = 0.11394
+
+[limit]
+output = 262_144
diff --git a/providers/maxlayer/models/thinkingmachines/inkling.toml b/providers/maxlayer/models/thinkingmachines/inkling.toml
new file mode 100644
index 00000000000..2756f6d9045
--- /dev/null
+++ b/providers/maxlayer/models/thinkingmachines/inkling.toml
@@ -0,0 +1,14 @@
+base_model = "thinkingmachines/inkling"
+structured_output = false
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "minimal", "low", "medium", "high", "max"]
+
+[cost]
+input = 1.08243
+output = 4.61457
+cache_read = 0.182304
+
+[limit]
+output = 262_144
diff --git a/providers/maxlayer/models/x-ai/grok-4.3.toml b/providers/maxlayer/models/x-ai/grok-4.3.toml
new file mode 100644
index 00000000000..6c8a30b4df2
--- /dev/null
+++ b/providers/maxlayer/models/x-ai/grok-4.3.toml
@@ -0,0 +1,13 @@
+base_model = "xai/grok-4.3"
+
+[[reasoning_options]]
+type = "effort"
+values = ["none", "low", "medium", "high"]
+
+[cost]
+input = 1.42425
+output = 2.8485
+cache_read = 0.22788
+
+[limit]
+output = 1_000_000
diff --git a/providers/maxlayer/models/x-ai/grok-4.5.toml b/providers/maxlayer/models/x-ai/grok-4.5.toml
new file mode 100644
index 00000000000..f4a3f91ae02
--- /dev/null
+++ b/providers/maxlayer/models/x-ai/grok-4.5.toml
@@ -0,0 +1,13 @@
+base_model = "xai/grok-4.5"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 2.2788
+output = 6.8364
+cache_read = 0.34182
+
+[modalities]
+input = ["text", "image", "pdf"]
diff --git a/providers/maxlayer/models/x-ai/grok-4.6.toml b/providers/maxlayer/models/x-ai/grok-4.6.toml
new file mode 100644
index 00000000000..613768c59a6
--- /dev/null
+++ b/providers/maxlayer/models/x-ai/grok-4.6.toml
@@ -0,0 +1,13 @@
+base_model = "xai/grok-4.6"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "xhigh"]
+
+[cost]
+input = 2.2788
+output = 6.8364
+cache_read = 0.5697
+
+[modalities]
+input = ["text", "image", "pdf"]
diff --git a/providers/maxlayer/models/x-ai/grok-build-0.1.toml b/providers/maxlayer/models/x-ai/grok-build-0.1.toml
new file mode 100644
index 00000000000..96f05c1706a
--- /dev/null
+++ b/providers/maxlayer/models/x-ai/grok-build-0.1.toml
@@ -0,0 +1,7 @@
+base_model = "xai/grok-build-0.1"
+reasoning_options = []
+
+[cost]
+input = 1.1394
+output = 2.2788
+cache_read = 0.22788
diff --git a/providers/maxlayer/models/xiaomi/mimo-v2.5-pro.toml b/providers/maxlayer/models/xiaomi/mimo-v2.5-pro.toml
new file mode 100644
index 00000000000..776c75a45ce
--- /dev/null
+++ b/providers/maxlayer/models/xiaomi/mimo-v2.5-pro.toml
@@ -0,0 +1,19 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "xiaomi/mimo-v2.5-pro"
+structured_output = true
+
+[interleaved]
+field = "reasoning_content"
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.495639
+output = 0.991278
+cache_read = 0.004102
+
+[limit]
+context = 1_050_000
diff --git a/providers/maxlayer/models/xiaomi/mimo-v2.5.toml b/providers/maxlayer/models/xiaomi/mimo-v2.5.toml
new file mode 100644
index 00000000000..561b1593ebc
--- /dev/null
+++ b/providers/maxlayer/models/xiaomi/mimo-v2.5.toml
@@ -0,0 +1,19 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "xiaomi/mimo-v2.5"
+structured_output = true
+
+[interleaved]
+field = "reasoning_details"
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.159516
+output = 0.319032
+cache_read = 0.003191
+
+[limit]
+context = 1_050_000
diff --git a/providers/maxlayer/models/z-ai/glm-4.5-air.toml b/providers/maxlayer/models/z-ai/glm-4.5-air.toml
new file mode 100644
index 00000000000..4d84f3a050c
--- /dev/null
+++ b/providers/maxlayer/models/z-ai/glm-4.5-air.toml
@@ -0,0 +1,13 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "zhipuai/glm-4.5-air"
+structured_output = false
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.148122
+output = 0.96849
+cache_read = 0.028485
diff --git a/providers/maxlayer/models/z-ai/glm-4.5.toml b/providers/maxlayer/models/z-ai/glm-4.5.toml
new file mode 100644
index 00000000000..8de1b16680e
--- /dev/null
+++ b/providers/maxlayer/models/z-ai/glm-4.5.toml
@@ -0,0 +1,13 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "zhipuai/glm-4.5"
+structured_output = false
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.68364
+output = 2.50668
+cache_read = 0.125334
diff --git a/providers/maxlayer/models/z-ai/glm-4.5v.toml b/providers/maxlayer/models/z-ai/glm-4.5v.toml
new file mode 100644
index 00000000000..adcebc9c691
--- /dev/null
+++ b/providers/maxlayer/models/z-ai/glm-4.5v.toml
@@ -0,0 +1,19 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "zhipuai/glm-4.5v"
+structured_output = false
+
+[[reasoning_options]]
+type = "toggle"
+
+[cost]
+input = 0.68364
+output = 2.05092
+cache_read = 0.125334
+
+[limit]
+context = 65_536
+
+[modalities]
+input = ["text", "image"]
diff --git a/providers/maxlayer/models/z-ai/glm-4.6.toml b/providers/maxlayer/models/z-ai/glm-4.6.toml
new file mode 100644
index 00000000000..5a014f5eb7f
--- /dev/null
+++ b/providers/maxlayer/models/z-ai/glm-4.6.toml
@@ -0,0 +1,8 @@
+base_model = "zhipuai/glm-4.6"
+structured_output = true
+reasoning_options = []
+
+[cost]
+input = 0.5697
+output = 2.2788
+cache_read = 0.11394
diff --git a/providers/maxlayer/models/z-ai/glm-4.6v.toml b/providers/maxlayer/models/z-ai/glm-4.6v.toml
new file mode 100644
index 00000000000..dbc433c679b
--- /dev/null
+++ b/providers/maxlayer/models/z-ai/glm-4.6v.toml
@@ -0,0 +1,11 @@
+base_model = "zhipuai/glm-4.6v"
+structured_output = false
+reasoning_options = []
+
+[cost]
+input = 0.34182
+output = 1.02546
+cache_read = 0.062667
+
+[limit]
+context = 131_072
diff --git a/providers/maxlayer/models/z-ai/glm-4.7-flash.toml b/providers/maxlayer/models/z-ai/glm-4.7-flash.toml
new file mode 100644
index 00000000000..b863d2a11c3
--- /dev/null
+++ b/providers/maxlayer/models/z-ai/glm-4.7-flash.toml
@@ -0,0 +1,15 @@
+base_model = "zhipuai/glm-4.7-flash"
+structured_output = true
+reasoning_options = []
+
+[interleaved]
+field = "reasoning_details"
+
+[cost]
+input = 0.068364
+output = 0.45576
+cache_read = 0.011394
+
+[limit]
+context = 202_752
+output = 16_384
diff --git a/providers/maxlayer/models/z-ai/glm-4.7.toml b/providers/maxlayer/models/z-ai/glm-4.7.toml
new file mode 100644
index 00000000000..d83e8e1e14b
--- /dev/null
+++ b/providers/maxlayer/models/z-ai/glm-4.7.toml
@@ -0,0 +1,11 @@
+base_model = "zhipuai/glm-4.7"
+structured_output = true
+reasoning_options = []
+
+[interleaved]
+field = "reasoning_details"
+
+[cost]
+input = 0.45576
+output = 1.99395
+cache_read = 0.091152
diff --git a/providers/maxlayer/models/z-ai/glm-5-turbo.toml b/providers/maxlayer/models/z-ai/glm-5-turbo.toml
new file mode 100644
index 00000000000..46cc2b24db8
--- /dev/null
+++ b/providers/maxlayer/models/z-ai/glm-5-turbo.toml
@@ -0,0 +1,14 @@
+base_model = "zhipuai/glm-5-turbo"
+structured_output = false
+reasoning_options = []
+
+[interleaved]
+field = "reasoning_content"
+
+[cost]
+input = 1.36728
+output = 4.5576
+cache_read = 0.273456
+
+[limit]
+context = 202_752
diff --git a/providers/maxlayer/models/z-ai/glm-5.1.toml b/providers/maxlayer/models/z-ai/glm-5.1.toml
new file mode 100644
index 00000000000..385145a2386
--- /dev/null
+++ b/providers/maxlayer/models/z-ai/glm-5.1.toml
@@ -0,0 +1,14 @@
+base_model = "zhipuai/glm-5.1"
+reasoning_options = []
+
+[interleaved]
+field = "reasoning_content"
+
+[cost]
+input = 1.100661
+output = 3.459219
+cache_read = 0.204409
+
+[limit]
+context = 204_800
+output = 128_000
diff --git a/providers/maxlayer/models/z-ai/glm-5.2.toml b/providers/maxlayer/models/z-ai/glm-5.2.toml
new file mode 100644
index 00000000000..d6ba7e9255d
--- /dev/null
+++ b/providers/maxlayer/models/z-ai/glm-5.2.toml
@@ -0,0 +1,22 @@
+# Toggle: reasoning.enabled = true|false
+# Effort: reasoning.effort (top-level reasoning_effort is an alias)
+# Budget: reasoning.max_tokens (integer reasoning tokens)
+base_model = "zhipuai/glm-5.2"
+
+[interleaved]
+field = "reasoning_content"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "effort"
+values = ["high", "xhigh"]
+
+[cost]
+input = 1.100661
+output = 3.459219
+cache_read = 0.220133
+
+[limit]
+context = 1_048_576
diff --git a/providers/maxlayer/models/z-ai/glm-5.3.toml b/providers/maxlayer/models/z-ai/glm-5.3.toml
new file mode 100644
index 00000000000..d046c5c3d0b
--- /dev/null
+++ b/providers/maxlayer/models/z-ai/glm-5.3.toml
@@ -0,0 +1,14 @@
+base_model = "zhipuai/glm-5.3"
+structured_output = false
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "high", "max"]
+
+[cost]
+input = 1.59516
+output = 5.01336
+cache_read = 0.296244
+
+[limit]
+context = 1_048_576
diff --git a/providers/maxlayer/models/z-ai/glm-5.toml b/providers/maxlayer/models/z-ai/glm-5.toml
new file mode 100644
index 00000000000..dadd9d7a0f8
--- /dev/null
+++ b/providers/maxlayer/models/z-ai/glm-5.toml
@@ -0,0 +1,14 @@
+base_model = "zhipuai/glm-5"
+structured_output = true
+reasoning_options = []
+
+[interleaved]
+field = "reasoning_content"
+
+[cost]
+input = 0.68364
+output = 2.187648
+cache_read = 0.136728
+
+[limit]
+output = 128_000
diff --git a/providers/maxlayer/models/z-ai/glm-5v-turbo.toml b/providers/maxlayer/models/z-ai/glm-5v-turbo.toml
new file mode 100644
index 00000000000..1cb7af4743e
--- /dev/null
+++ b/providers/maxlayer/models/z-ai/glm-5v-turbo.toml
@@ -0,0 +1,14 @@
+base_model = "zhipuai/glm-5v-turbo"
+structured_output = false
+reasoning_options = []
+
+[cost]
+input = 1.36728
+output = 4.5576
+cache_read = 0.273456
+
+[limit]
+context = 202_752
+
+[modalities]
+input = ["image", "text", "video"]
diff --git a/providers/maxlayer/provider.toml b/providers/maxlayer/provider.toml
new file mode 100644
index 00000000000..b95b0765fbb
--- /dev/null
+++ b/providers/maxlayer/provider.toml
@@ -0,0 +1,5 @@
+name = "Maxlayer"
+env = ["MAXLAYER_API_KEY"]
+npm = "@ai-sdk/openai-compatible"
+api = "https://inference.maxlayer.cloud/v1"
+doc = "https://maxlayer.cloud/docs/inference"