catalog: add backend:cuda entries for the 0.2.0 CUDA release - #34
Merged
Conversation
Sync the 22 CUDA-native catalog entries (11 models x cuda-q4mix + cuda-q8) from the engine repo. The runtime fetches this catalog from public main (raw.githubusercontent.com/basecompute/baseRT/main/.../catalog.json), and that hosted copy takes precedence over the binary's bundled fallback — so without these entries a CUDA (GB10) user on 0.2.0 cannot resolve the cuda-native bf16- scale bundles (f16-scale universal bundles don't load on CUDA). The .base bundles are already public on HF. Non-CUDA entries are unchanged (byte-identical).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The runtime resolves models by fetching this catalog from public
main(raw.githubusercontent.com/basecompute/baseRT/main/base-convert/crates/base-hub/catalog.json), and that hosted copy takes precedence over the binary's bundled fallback (resolution order: fresh cache → hosted fetch → stale cache → bundled).Today public
catalog.jsonhas 0backend:cudaentries, so a CUDA (GB10) user on the upcoming 0.2.0 release cannot resolve the CUDA-native bundles — the whole CUDA release would be invisible to them. This adds the 22 entries (11 models ×cuda-q4mix+cuda-q8) synced verbatim from the engine repo..basebundles are already public on HF (spot-checked HTTP 302).v0.2.0tag so CUDA resolution works at runtime.Pairs with #33 (install.sh Linux/CUDA support).