fix: CUDA→CPU whisper fallback via Foundry variant API#18
Merged
Conversation
The previous fallback used string manipulation on model IDs and catalog.GetModelAsync, which always resolves to the GPU variant. Use model.Variants to find the CPU variant and catalog.GetModelVariantAsync to load it — the correct Foundry Local API for variant-specific resolution. Fixes #17 See also microsoft/Foundry-Local#567 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
cudnn_engines_precompiled64_9.dllin Foundry Local's CUDA EP bundlemodel.Variants+catalog.GetModelVariantAsync()instead of string-manipulated IDs withcatalog.GetModelAsync()(which always resolves to the GPU variant)IsCudaError) — catches cuDNN/CUDA/OnnxRuntimeGenAI errors on first transcription attempt and transparently retries on CPUContext
Foundry Local's CUDA EP bundle (v12.8.2) is missing
cudnn_engines_precompiled64_9.dll, causing all Whisper CUDA variants to fail at inference withCUDNN_BACKEND_API_FAILEDon the encoder's FP16 Conv node. Filed upstream as microsoft/Foundry-Local#567.The previous fallback used
catalog.GetModelAsync(cpuId)with string-replaced model IDs — this API only resolves aliases and always returns the GPU variant, so the fallback could never find the CPU model.Fixes #17
Test plan
catalog.GetModelVariantAsync()+ModelVariant.DownloadAsync/LoadAsync/GetAudioClientAsyncaspire start— all resources healthy, Telegram service running with fix🤖 Generated with Claude Code