Skip to content

Promote to production: voice-to-text dtype fix - #73

Merged
slaveofcode merged 2 commits into
mainfrom
develop
Aug 1, 2026
Merged

Promote to production: voice-to-text dtype fix#73
slaveofcode merged 2 commits into
mainfrom
develop

Conversation

@slaveofcode

Copy link
Copy Markdown
Owner

Fixes the Whisper 'Can't create a session' (MatMulNBits missing scale) error via per-module dtype.

🤖 Generated with Claude Code

Kresna and others added 2 commits August 1, 2026 13:01
Reported: 'Can't create a session … MatMulNBits Missing required scale:
model.decoder.embed_tokens.weight … DequantizeLinear'.

Whisper is an encoder-decoder model. A single-string dtype ('q8', the WASM
default) left the decoder — which holds embed_tokens — on a 4-bit (MatMulNBits)
variant whose scale tensor the bundled ONNX Runtime can't resolve, so the
session fails to create. Switch to a per-module dtype and force
decoder_model_merged to fp32 (unquantized, proper scales); the encoder stays
q8 on WASM / fp32 on WebGPU to limit the download.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
fix(voice-to-text): per-module dtype so the Whisper session loads
@slaveofcode
slaveofcode merged commit d14bfe6 into main Aug 1, 2026
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant