v1.17.0

xorbitsai/inferencev1.17.0Jan 10, 2026by XprobeBot

AI Summary

Version 1.17.0 adds enable_thinking argument support, MThreads (MUSA) GPU support, minimax tool call support, and new Qwen image models. Includes FP4 quantization support, video GGUF support, and multi-engine configurations.

Key Highlights

  • enable_thinking argument support for models
  • MThreads (MUSA) GPU support
  • Minimax tool call support
  • Qwen-Image-Edit-2512 and Qwen-Image-2512 models
  • FP4 quantization support
  • Video GGUF support
  • Multi-engine support for OCR, DeepSeek OCR, and image models
  • Distributed model launch for vLLM >= v0.11.0

New Features

  • enable_thinking
  • MThreads (MUSA) GPU
  • minimax tool call
  • Qwen-Image-Edit-2511
  • Qwen-Image-2512
  • FP4 support
  • video gguf
  • multi engines for ocr
  • multi engines for deepseek ocr
  • multi engines for image models

Full Release Notes

# What's new in 1.17.0 (2026-01-10)
These are the changes in inference v1.17.0.
## New features
* FEAT: add enable_thinking kwarg support by @OliverBryant in https://github.com/xorbitsai/inference/pull/4423
* FEAT: Support MThreads (MUSA) GPU by @yeahdongcn in https://github.com/xorbitsai/inference/pull/4425
* FEAT: support distributed model launch for vllm version>=v0.11.0 by @OliverBryant in https://github.com/xorbitsai/inference/pull/4428
* FEAT: [model] Qwen-Image-Edit-2511 support by @OliverBryant in https://github.com/xorbitsai/inference/pull/4427
* FEAT: add minimax tool call support by @OliverBryant in https://github.com/xorbitsai/inference/pull/4434
* FEAT: [model] Qwen-Image-2512 support by @OliverBryant in https://github.com/xorbitsai/inference/pull/4435
* FEAT: support auto batch for sentence_transformers rerank by @llyycchhee in https://github.com/xorbitsai/inference/pull/4429
* FEAT: add multi engines for ocr && deepseek ocr mlx support by @OliverBryant in https://github.com/xorbitsai/inference/pull/4437
* FEAT: add fp4 support by @OliverBryant in https://github.com/xorbitsai/inference/pull/4450
* FEAT: add video gguf support by @OliverBryant in https://github.com/xorbitsai/inference/pull/4458
* FEAT: add multi engines for image model by @OliverBryant in https://github.com/xorbitsai/inference/pull/4446
## Enhancements
* ENH: update 4 models JSON ("Deepseek-V3.1", "deepseek-r1-0528", "deepseek-r1-0528-qwen3", ... +1 more) by @OliverBryant in https://github.com/xorbitsai/inference/pull/4445
* ENH: update model "DeepSeek-OCR" JSON by @OliverBryant in https://github.com/xorbitsai/inference/pull/4444
* ENH:  support vllm mtp & rope scaling by @ZhikaiGuo960110 in https://github.com/xorbitsai/inference/pull/4454
## Bug fixes
* BUG: fix empty cache for vllm embedding & rerank by @ZhikaiGuo960110 in https://github.com/xorbitsai/inference/pull/4422
* BUG: Selecting the same worker repeatedly by @OliverBryant in https://github.com/xorbitsai/inference/pull/4447
* BUG: fix vllm ocr model cannot stop by @OliverBryant in https://github.com/xorbitsai/inference/pull/4460
* BUG: Models being downloaded cannot be canceled. by @OliverBryant in https://github.com/xorbitsai/inference/pull/4461
## Documentation
* DOC: update new models and release notes for v1.16.0 by @qinxuye in https://github.com/xorbitsai/inference/pull/4416
* DOC: update docker docs by @qinxuye in https://github.com/xorbitsai/inference/pull/4419
* DOC:  vLLM + Torch + Xinference Compatibility Issue by @qiulang in https://github.com/xorbitsai/inference/pull/4442

## New Contributors
* @yeahdongcn made their first contribution in https://github.com/xorbitsai/inference/pull/4425

**Full Changelog**: https://github.com/xorbitsai/inference/compare/v1.16.0...v1.17.0