github AlexsJones/llmfit v1.1.3

8 hours ago

1.1.3 (2026-07-14)

Features

  • add NCAI VAETKI model to curated model database (bdbf338)
  • add RamaLama as a detection provider (5bef0b2)

Bug Fixes

  • hardware: read discrete Intel Arc VRAM from sysfs (xe/i915) (9bbaa8d)
  • hardware: read discrete Intel Arc VRAM from sysfs (xe/i915) (9fbbcb8), closes #609
  • match Unsloth -UD dynamic GGUFs to catalog models (b3de39d)
  • match Unsloth -UD dynamic GGUFs to catalog models (e9b66be)
  • name AMD unified APUs when rocm-smi reports "N/A", gate junk shares (ba96f4c)
  • name AMD unified APUs when rocm-smi reports "N/A", gate junk shares (e282011)
  • providers: detect oMLX as an MLX runtime (f649952)
  • providers: show LM Studio/Docker Desktop as installed when idle (bca491e)
  • providers: show LM Studio/Docker Desktop as installed when idle (5a22add), closes #731
  • serve: include per-GPU memory_bandwidth_gbps in /api/v1/system (ce3f7bf)
  • serve: include per-GPU memory_bandwidth_gbps in /api/v1/system (8b9b7aa), closes #747

What's Changed

  • fix(providers): detect oMLX as an MLX runtime by @CN-Scars in #701
  • feat: add RamaLama as a detection provider by @Olawoyin007 in #700
  • fix: match Unsloth -UD dynamic GGUFs to catalog models by @AlexsJones in #729
  • fix(providers): show LM Studio/Docker Desktop as installed when idle by @AlexsJones in #734
  • feat: add NCAI VAETKI model to curated model database by @wklee610 in #658
  • fix(hardware): read discrete Intel Arc VRAM from sysfs (xe/i915) by @AlexsJones in #733
  • chore(deps): bump assert_cmd from 2.2.1 to 2.2.2 by @dependabot[bot] in #579
  • chore(deps): bump unicode-segmentation from 1.13.2 to 1.13.3 by @dependabot[bot] in #646
  • chore(deps): bump actions/cache from 5.0.5 to 6.1.0 by @dependabot[bot] in #694
  • fix: name AMD unified APUs when rocm-smi reports "N/A", gate junk shares by @AlexsJones in #742
  • Correct MiniMax M2.7 context window by @octo-patch in #743
  • chore(deps): bump docker/metadata-action from 6.0.0 to 6.2.0 by @dependabot[bot] in #741
  • chore(deps): bump serde_json from 1.0.149 to 1.0.150 by @dependabot[bot] in #595
  • fix(serve): include per-GPU memory_bandwidth_gbps in /api/v1/system by @AlexsJones in #749
  • chore(main): release 1.1.3 by @AlexsJones in #727

New Contributors

Full Changelog: v1.1.2...v1.1.3

Don't miss a new llmfit release

NewReleases is sending notifications on new releases.