v0.3.9

What's Changed

• Fix qwen3_vl ValueError: Image features and image tokens do not match by @ziya32 in #608
• Add ministral3 by @Blaizzy in #611

New Contributors

• @ziya32 made their first contribution in #608

Full Changelog: v0.3.8...v0.3.9

v0.3.8

What's Changed

• Add chat_ui command by @Blaizzy in #589
• Fix 422 error when calling /chat/completions with OpenAI SDK #590 by @Blaizzy in #592
• Fix qwen3_vl by @awni in #599

Full Changelog: v0.3.7...v0.3.8

v0.3.7

What's Changed

• update readme with new openai endpoints details by @mguella in #585
• Fix lighton OCR empty result on main by @Blaizzy in #587
• Add model revision loading by @Blaizzy in #588

Full Changelog: v0.3.6...v0.3.7

v0.3.6

What's Changed

• Fix: Input cast error by @Blaizzy in #559
• Fix qwen-vl position ids (2.5 & 3) by @Blaizzy in #564
• Add Evals by @Blaizzy in #563
• Fix Qwen3-VL Attention by @Blaizzy in #566
• Update evals init by @Blaizzy in #567
• Fix Qwen3-VL multi image reshape by @Blaizzy in #569
• Add processor args to DeepSeekOCR by @Blaizzy in #570
• host and port params for server by @mguella in #568
• FastVLM by @pcuenca in #502
• Add support for InternVL3 by @iRonJ in #540
• Add Z.ai GLM-4.1v by @Blaizzy in #572
• Make rope deltas private by @Blaizzy in #573
• Add example notebook for interleaving text and images in prompts by @Copilot in #574
• [Bugfix] fix mrope in qwen2vl and qwen2.5vl by @JJJYmmm in #576
• Add lighton-ocr by @Blaizzy in #550
• changed image parameter instead of files in stream_generate by @Manikandan-t in #521
• Remove auto config loading by @Blaizzy in #577
• [BugFix][Qwen3VL] fix deepstack and multi-image inference by @JJJYmmm in #581
• openai compatible endpoints by @mguella in #580
• Bump version to 0.3.6 by @Blaizzy in #582

New Contributors

• @mguella made their first contribution in #568
• @iRonJ made their first contribution in #540
• @Copilot made their first contribution in #574
• @JJJYmmm made their first contribution in #576
• @Manikandan-t made their first contribution in #521

Full Changelog: v0.3.5...v0.3.6

v0.3.5

What's Changed

• Remove docs actions temporarly by @Blaizzy in #536
• Fix load_image for URLs by @pcuenca in #534
• Add Deepseek ocr by @Blaizzy in #541

Full Changelog: v0.3.4...v0.3.5

v0.3.4

What's Changed

• Add n_kv_heads property used in LM Studio to glm4v_moe.LanguageModel by @hehua2008 in #472
• Fix: rope rotation (GLM-4.5v) by @Blaizzy in #481
• Remove scipy dep by @Blaizzy in #482
• Add CUDA and CPU as optional deps by @Blaizzy in #483
• Fix Deepseek vl2 chat template by @Blaizzy in #488
• Fix deepseek-vl default chat template by @Blaizzy in #490
• Fix smolvlm video generate by @Blaizzy in #491
• Fix base64 encoded images by @Blaizzy in #493
• Map Apriel configs to pixtral and fix prompt formatting by @ivanfioravanti in #518
• Fix video understanding NB by @Blaizzy in #519
• Fix fine-tuning bug in trainer.py by @avishekjana in #473
• Bump minimum required Python version to 3.10 by @dokterbob in #485
• Add Qwen3-VL (Dense & MoE) by @Blaizzy & @vincentamato in #528
• Fix video Qwen3-VL by @Blaizzy in #529
• Fix Qwen3 VL (Dense) Sanitize by @Blaizzy in #531
• Bump version to 0.3.4 by @Blaizzy in #535

New Contributors

• @hehua2008 made their first contribution in #472
• @ivanfioravanti made their first contribution in #518
• @avishekjana made their first contribution in #473
• @dokterbob made their first contribution in #485
• @vincentamato made their first contributuin in #529

Full Changelog: v0.3.3...v0.3.4

v0.3.3

What's Changed

• fix changelog task by @Blaizzy in #441
• Add LFM2-VL by @Blaizzy in #460
• [llava_next] Fix config inheritance by @neilmehta24 in #448
• Kimi_VL: Fix activation args by @Blaizzy in #465
• Add GLM-4-5V by @Blaizzy in #458
• External access to LFM2-VL merge-input-IDs method by @christian-lms in #466
• Add Command-A-Vision by @Blaizzy in #467
• [kernels] Use a header for bicubic_interpolate for compatibility with macOS < 15 by @mattjcly in #469

New Contributors

• @christian-lms made their first contribution in #466

Full Changelog: v0.3.2...v0.3.3

v0.3.2

What's Changed

• Fix energy calc in omni.py by @Blaizzy in #427
• Feat(build): migrate to pyproject.toml by @SauravMaheshkar in #282
• Fix quant predicate by @Blaizzy in #430
• Load dependencies from requirements.txt by @Blaizzy in #431
• Fix broken wheel builds caused by ambiguous package spec by @neilmehta24 in #432
• Make UI and audio dependencies optional by @zhnext in #433
• Cleanup: Refactor config by @Blaizzy in #437
• Add cuda support by @Blaizzy in #438
• Fix/server module exposure by @zhnext in #434
• Support text only training by @Goekdeniz-Guelmez in #424
• Fix phi3_v and molmo mask by @Blaizzy in #440

New Contributors

• @zhnext made their first contribution in #433
• @Goekdeniz-Guelmez made their first contribution in #424

Full Changelog: v0.3.1...v0.3.2

v0.3.1

What's Changed

• fix(chat-ui): Fix imports, blocking Chat-ui server start by @zenyr in #420
• Fix server empty image for Gemma3n by @Blaizzy in #422
• [Gemma3n] Add hooks for image embedding computation by @will-lms in #407
• [gemma3n] Fix OCR after weight re-upload by @neilmehta24 in #425
• Add gemma3n omni example by @Blaizzy in #426

New Contributors

• @zenyr made their first contribution in #420

Full Changelog: v0.3.0...v0.3.1

v0.3.0

What's Changed

• [gemma3n] Correctly scale text embeddings for quantized gemma3n conversions by @neilmehta24 in #397
• smolvlm_video example: fix typo in system prompt by @pcuenca in #389
• Fix gemma3n pixel casting by @Blaizzy in #398
• Fix audio model check and prompt utils by @Blaizzy in #395
• Add KV Quantization by @Blaizzy in #401
• Fix Gemma3n multi-task merging and update LM by @Blaizzy in #405
• [gemma3n] Fix vision encoder implementation of EdgeResidual and UniversalInvertedResidual by @neilmehta24 in #410
• fix: Remove unnecessary unicode_escape decoding for Chinese text input by @nicekate in #403
• Add support for Mixed Quant by @Blaizzy in #413
• Fix gemma3n Vision OCR + LM only reponses by @Blaizzy in #414
• Fix generate signature by @Blaizzy in #416
• Add support for audio modality in server by @Blaizzy in #417
• Update server, readme and misc by @Blaizzy in #418

New Contributors

• @nicekate made their first contribution in #403

Full Changelog: v0.2.0...v0.3.0