feat: add Avian as a cloud LLM inference provider

[avianion]

Summary

Adds Avian as a new Go backend for LocalAI, enabling users to access cloud-hosted LLMs through Avian's OpenAI-compatible API at https://api.avian.io/v1.

Backend

• New Go gRPC backend at backend/go/avian/ following the existing huggingface backend pattern
• Proxies chat completion requests (both streaming and non-streaming) to the Avian API
• Supports structured messages, temperature, top_p, max_tokens, and stop sequences
• Authentication via AVIAN_API_KEY environment variable
• Configurable base URL via AVIAN_API_BASE environment variable (defaults to https://api.avian.io/v1)

Gallery Models

Four models available out of the box:

Model	Context	Output	Input Price	Output Price
deepseek/deepseek-v3.2	164K	65K	$0.26/1M	$0.38/1M
moonshotai/kimi-k2.5	131K	8K	$0.45/1M	$2.20/1M
z-ai/glm-5	131K	16K	$0.30/1M	$2.55/1M
minimax/minimax-m2.5	1M	1M	$0.30/1M	$1.10/1M

Usage

# Set API key
export AVIAN_API_KEY=your-api-key

# Install the model from the gallery
local-ai models install avian-deepseek-v3.2

# Or use a model config YAML
cat > models/avian-deepseek.yaml << EOF
name: avian-deepseek
backend: avian
parameters:
  model: deepseek/deepseek-v3.2
  max_tokens: 8192
context_size: 164000
EOF

# Query the model
curl http://localhost:8080/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model":"avian-deepseek-v3.2","messages":[{"role":"user","content":"Hello!"}]}'

Build Infrastructure

• Backend definition in Makefile (BACKEND_AVIAN = avian|golang|.|false|true)
• CI workflow entries for Linux (amd64/arm64) and macOS (metal/darwin)
• Backend index.yaml entries with OCI image references
• Added to .NOTPARALLEL and docker-build-backends targets

Files Changed

• backend/go/avian/ - Go gRPC backend implementation (main.go, avian.go, Makefile, run.sh, package.sh)
• gallery/avian.yaml - Base model configuration template
• gallery/index.yaml - Four Avian model entries
• backend/index.yaml - Backend metadata and OCI image entries
• Makefile - Backend build targets
• .github/workflows/backend.yml - CI build matrix entries

Test plan

• Backend compiles: make -C backend/go/avian
• Docker build succeeds: BACKEND=avian make docker-build-avian
• Model loads with valid AVIAN_API_KEY
• Chat completions work (non-streaming)
• Chat completions work (streaming)
• Gallery model install works: local-ai models install avian-deepseek-v3.2
• Error handling: missing API key returns clear error message

cc @mudler

[netlify]

✅ Deploy Preview for localai ready!

Name	Link
🔨 Latest commit	023abe2
🔍 Latest deploy log	https://app.netlify.com/projects/localai/deploys/69ab42c401003b0008f6c3d6
😎 Deploy Preview	https://deploy-preview-8666--localai.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify project configuration.

[avianion]

Hey @mudler, would love your review on this when you get a chance. Happy to address any feedback!

[avianion]

Friendly follow-up — this PR is still active and ready for review. Would appreciate a look when you get a chance! cc @mudler

[avianion]

Friendly follow-up — this PR is still active and ready for review. All feedback has been addressed. Would appreciate a look when you get a chance! cc @mudler

[avianion]

Hey @mudler — friendly follow-up on this PR. Avian is an OpenAI-compatible inference provider that's already live and powering apps like ISEKAI ZERO. This is a lightweight integration (standard OpenAI-compatible endpoint) and we're happy to address any feedback or make adjustments. Would love to get this merged if you have a moment to review. Thanks!

[localai-bot]

Hello! Thank you for your contribution. Before we can proceed with this PR, we need to have a discussion about the proposed changes.

LocalAI doesn't support cloud-based backends, and there was no issue opened for this feature prior to the PR. We encourage contributors to first open an issue to discuss their proposed changes with the maintainers.

Could you please open an issue describing your use case and proposed implementation? Once we've had a chance to discuss and align on the direction, we can then continue reviewing this PR.

Thank you for your understanding!

[avianion]

Hi @mudler — thank you for the feedback. As requested, I've opened a discussion issue at #8877 to discuss whether LocalAI should support cloud-based inference providers like Avian.

Happy to provide any additional information or adjust the approach based on the maintainers' direction. Please feel free to review the issue and close this PR if the discussion determines cloud backends are out of scope — or let me know if there are specific changes needed to make this PR viable.