xenova/transformers.js

Pull Requests Commits

[version] Update to 4.1.0

xenova committed 39 days ago

2af2d505

bump onnxruntime-web

xenova committed 39 days ago

c0e0d3b6

xenova committed 39 days ago

5419e31c

Update default generation parameters (#1649)

xenova committed 39 days ago

Verified fbab88f6

Re-enable SmolVLM (#1648)

pdufour committed 39 days ago

Verified 39bec702

chore: bump doc-builder SHA for PR upload workflow (#1644)

rtrompier committed 39 days ago

Verified 78d4ec3d

Add support for q1, q1f16, q2, q2f16 data types (#1647)

xenova committed 40 days ago

Verified fdd496e9

Improve tokenizer types based on input function parameters (#1641)

xenova committed 41 days ago

Verified ed84c15f

Cached generation improvements (+ past_key_values via pipeline function) (#1638)

xenova committed 41 days ago

Verified 8d615c5a

🔒 Pin GitHub Actions to commit SHAs (#1626)

paulinebm committed 47 days ago

Verified 60379e22

[version] Update to 4.0.1

xenova committed 52 days ago

Verified 64da6359

casually dropping the most capable open weights on the planet (#1627)

xenova committed 52 days ago

Verified 3cee1ac2

[version] Update to 4.0.0 (#1620)

xenova committed 55 days ago

Verified 364ebd41

[version] Update to 4.0.0-next.11

xenova committed 56 days ago

Verified 3458fa5e

Upgrade onnxruntime-web version (#1618)

xenova committed 56 days ago

Verified 28f92674

Fix custom cache in Node.js when cache returns non-paths (e.g., Response) (#1617)

xenova committed 56 days ago

Verified ab22b27e

Emit `progress_total` events from `PreTrainedModel.from_pretrained()` (#1615)

xenova committed 56 days ago

Verified d8340931

Fix ModelRegistry calls when loading multimodal models with text-generation pipeline (#1614)

xenova committed 56 days ago

Verified a0d86d50

Overdue Whisper fixes (#1594)

xenova committed 56 days ago

Verified 43b26629

[version] Update to 4.0.0-next.10

xenova committed 58 days ago

Verified d6280f73

Add support for Cohere ASR (#1610)

xenova committed 58 days ago

Verified b368d75e

Fix typo in WebGPU documentation (#1607)

mattsoulanille committed 60 days ago

Verified e94d5592

[version] Update to 4.0.0-next.9

xenova committed 61 days ago

Verified d5610ad6

feat: add ModelRegistry.get_available_dtypes() to detect quantization (#1580)

sroussey committed 62 days ago

Verified b8228dad

Add support for qwen3_5_text <-> Qwen3_5ForCausalLM (#1602)

xenova committed 62 days ago

Verified 99a8eebc

Add support for Gemma3 VLM architecture (#1601)

xenova committed 62 days ago

Verified a97b51bc

[version] Update to 4.0.0-next.8

xenova committed 67 days ago

Verified 330021ca

Add background-removal pipeline to JSDoc (#1595)

xenova committed 68 days ago

Verified 3cfb8dd3

Add support for SolarOpen and CHMv2 models (#1593)

xenova committed 68 days ago

Verified f169789c

Align audio_spectrogram_transformer name with python library (#1591)

xenova committed 68 days ago

Verified 8a905cd6

Older