oMLX

最近の更新

• v0.7.2 2026年10月21日
  リリースノート（原文）

  # oMLX 0.7.2
  
  oMLX 0.7.2 brings faster DeepSeek V4.1 on every oQ build, Lightning MTP across Macs in distributed inference, and Qwen3.8-Flash-Next GGUFs on Splash. It's only been a day since 0.7.1, but these were worth shipping now.
  
  **Download:** [macOS 26 / 27](https://github.com/jundot/omlx/releases/download/v0.7.2/oMLX-0.7.2-macos26-27.dmg) | [macOS 15 Sequoia](https://github.com/jundot/omlx/releases/download/v0.7.2/oMLX-0.7.2-macos15-sequoia.dmg)
  
  ## Faster DeepSeek V4.1 on oQ2e to oQ4e
  
  The fused decode kernels in 0.7.1 only read the MXFP weights of oQ4e, so oQ2e, oQ2.5e and oQ3e still decoded on a slower generic path. They now run on the fused kernels too, Lightning MTP verify shares work across the checked tokens, and prefill runs in 8192-token chunks. #4722
  
  > DeepSeek-V4.1-Flash-…

• v0.7.1 2026年10月20日
  リリースノート（原文）

  # oMLX 0.7.1
  
  oMLX 0.7.1 adds GGUF downloads served through Splash, much faster DeepSeek V4.1 decode, 4-bit and 8-bit KV cache formats, a fix for GPU hangs when several models run at once, and distributed inference for more models. The notes below cover changes since 0.7.1rc1, and earlier 0.7.1 highlights are summarized at the end.
  
  **Download:** [macOS 26 / 27](https://github.com/jundot/omlx/releases/download/v0.7.1/oMLX-0.7.1-macos26-27.dmg) | [macOS 15 Sequoia](https://github.com/jundot/omlx/releases/download/v0.7.1/oMLX-0.7.1-macos15-sequoia.dmg)
  
  ## GGUF Models Through Splash
  
  The model downloader now filters Hugging Face repos by MLX or GGUF and lets you pick the files you want, and a GGUF variant brings its vision projector along. Downloaded GGUFs show up as models, and Qwen3.8-27B …

• v0.7.1rc1 2026年10月15日
  リリースノート（原文）

  # oMLX 0.7.1rc1
  
  > This is a release candidate. Please report any bugs through [GitHub Issues](https://github.com/jundot/omlx/issues/new/choose). If you need a stable release, please use [oMLX 0.7.0](https://github.com/jundot/omlx/releases/tag/v0.7.0).
  
  The release candidate for oMLX 0.7.1 adds Muse Spark Open and brings everything from the 0.7.1 development releases: model recommendations for your Mac, Splash as an external engine, reworked model settings and chat, DFlash in the batched engines, decision models and batched prefill.
  
  **Download:** [macOS 26 / 27](https://github.com/jundot/omlx/releases/download/v0.7.1rc1/oMLX-0.7.1rc1-macos26-27.dmg) | [macOS 15 Sequoia](https://github.com/jundot/omlx/releases/download/v0.7.1rc1/oMLX-0.7.1rc1-macos15-sequoia.dmg)
  
  ## Muse Spark Ope…