Commits

Branch selector

main

User selector

All users

All time

Commit history

Commits on Apr 2, 2026

• Remove TurboQuant benchmark artifacts and add README docs (#894)

  Show description for 391179c
  

  Blaizzy

  and

  claude

  authored
  391179cView commit details

  Copy full SHA for 391179c

  Browse repository at this point

• Add Turbo Quant (#858)

  Show description for 3f133c7
  

  Blaizzy

  and

  claude

  authored
  3f133c7View commit details

  Copy full SHA for 3f133c7

  Browse repository at this point

• Fix Gemma 4 embedding scaling (#893)

  Show description for 9db27b5
  

  Blaizzy

  and

  claude

  authored
  9db27b5View commit details

  Copy full SHA for 9db27b5

  Browse repository at this point

• Add Gemma 4 model support (vision, audio, MoE) (#890)

  Show description for 756dd89
  

  Blaizzy

  and

  pcuenca

  authored
  756dd89View commit details

  Copy full SHA for 756dd89

  Browse repository at this point

• fix(server): disable uvicorn reload by default to prevent memory leaks (#883)

  Show description for a0a0663
  

  3 peopleauthored
  a0a0663View commit details

  Copy full SHA for a0a0663

  Browse repository at this point

• Add falcon perception (#888)

  Show description for f6cdc9e
  Blaizzy

  authored
  f6cdc9eView commit details

  Copy full SHA for f6cdc9e

  Browse repository at this point

Commits on Apr 1, 2026

• Add wired_limit to vision model inference pipelines for CUDA support (#887)

  Show description for c770022
  

  Blaizzy

  and

  claude

  authored
  c770022View commit details

  Copy full SHA for c770022

  Browse repository at this point

• Add granite vision 3.2 and 4.0 (#885)

  Show description for 1a1820d
  Blaizzy

  authored
  1a1820dView commit details

  Copy full SHA for 1a1820d

  Browse repository at this point

Commits on Mar 31, 2026

• Add RF-DETR detection and segmentation model (#884)

  Show description for 590647f
  

  Blaizzy

  and

  claude

  authored
  590647fView commit details

  Copy full SHA for 590647f

  Browse repository at this point

Commits on Mar 30, 2026

• Add Falcon-OCR model support (#879)

  Show description for 1f0c0ff
  

  4 peopleauthored
  1f0c0ffView commit details

  Copy full SHA for 1f0c0ff

  Browse repository at this point

• Add SAM 3.1 with Object Multiplex and optimized realtime pipeline (#880)

  Show description for ab58386
  

  Blaizzy

  and

  claude

  authored
  ab58386View commit details

  Copy full SHA for ab58386

  Browse repository at this point

Commits on Mar 28, 2026

• Fix PaliGemma processor kwarg routing (#877)

  Show description for 7f7a04c
  Blaizzy

  authored
  7f7a04cView commit details

  Copy full SHA for 7f7a04c

  Browse repository at this point

• Add DOTS-MOCR processor support (#874)

  Show description for 3e06f96
  Blaizzy

  authored
  3e06f96View commit details

  Copy full SHA for 3e06f96

  Browse repository at this point

• [Sam3] label drawing for mask only in realtime videos (#876)

  Show description for 673986e
  Blaizzy

  authored
  673986eView commit details

  Copy full SHA for 673986e

  Browse repository at this point

• Add Facebook Sam3 (#875)

  Show description for 01f6e8e
  Blaizzy

  authored
  01f6e8eView commit details

  Copy full SHA for 01f6e8e

  Browse repository at this point

• Fix Magistral image token expansion (#873)

  Show description for 3426d3a
  Blaizzy

  authored
  3426d3aView commit details

  Copy full SHA for 3426d3a

  Browse repository at this point

• Fix LFM2-VL processor loading without torch (#872)

  Show description for 4b868a6
  Blaizzy

  authored
  4b868a6View commit details

  Copy full SHA for 4b868a6

  Browse repository at this point

Commits on Mar 27, 2026

• Fix thinking defaults in CLI and server (#869)

  Show description for 28ce554
  Blaizzy

  authored
  28ce554View commit details

  Copy full SHA for 28ce554

  Browse repository at this point

• Cast Qwen3.5 RMSNorm output to input dtype (#868)

  Blaizzy

  authored
  a63714eView commit details

  Copy full SHA for a63714e

  Browse repository at this point

• Patch qwen3_5_moe auto processor (#867)

  Show description for 094adac
  

  Blaizzy

  and

  joncooper

  authored
  094adacView commit details

  Copy full SHA for 094adac

  Browse repository at this point

• fix(qwen3_5): register AutoProcessor patch for qwen3_5 model type (#859)

  Show description for e4923d8
  

  mdstaff

  and

  Blaizzy

  authored
  e4923d8View commit details

  Copy full SHA for e4923d8

  Browse repository at this point

• Ensure minimal metadata exists in model card (#853)

  Show description for 124fb28
  pcuenca

  authored
  124fb28View commit details

  Copy full SHA for 124fb28

  Browse repository at this point

Commits on Mar 21, 2026

• Add molmo point (#844)

  Blaizzy

  authored
  f613272View commit details

  Copy full SHA for f613272

  Browse repository at this point

• Set custom processor as default and remove torch as dep (#821)

  Show description for ca5e1f2
  

  Blaizzy

  and

  claude

  authored
  ca5e1f2View commit details

  Copy full SHA for ca5e1f2

  Browse repository at this point

Commits on Mar 19, 2026

• Clean up server + generate args, validation, and defaults (#829)

  Show description for a8f9803
  

  spicyneuron

  and

  Blaizzy

  authored
  a8f9803View commit details

  Copy full SHA for a8f9803

  Browse repository at this point

Commits on Mar 17, 2026

• Add mistral4 (#827)

  Show description for 822a843
  Blaizzy

  authored
  822a843View commit details

  Copy full SHA for 822a843

  Browse repository at this point

Commits on Mar 14, 2026

• Fix Qwen3-Omni (qwen3_omni_moe) integration (#820)

  Show description for b7f853a
  

  3 peopleauthored
  b7f853aView commit details

  Copy full SHA for b7f853a

  Browse repository at this point

• Add an 'index' for each tool call as per OpenAI spec. (#818)

  viktike

  authored
  fb70686View commit details

  Copy full SHA for fb70686

  Browse repository at this point

• Add common sampling parameters to server endpoints (#814)

  Show description for c9bdefb
  spicyneuron

  authored
  c9bdefbView commit details

  Copy full SHA for c9bdefb

  Browse repository at this point

Commits on Mar 11, 2026

• fix(qwen3_5): guard batch dimension mismatches under continuous batching (#813)

  Show description for db3d558
  kol22

  authored
  db3d558View commit details

  Copy full SHA for db3d558

  Browse repository at this point

• feat(server): add --model and --adapter-path flags for startup preloading (#811)

  Show description for 08e6bc5
  auggie246

  authored
  08e6bc5View commit details

  Copy full SHA for 08e6bc5

  Browse repository at this point

Commits on Mar 7, 2026

• Update dependencies and version number (#809)

  Show description for 718f69e
  Blaizzy

  authored
  718f69eView commit details

  Copy full SHA for 718f69e

  Browse repository at this point

• Remove sleep duration with zero delay (#808)

  Show description for bbb696c
  Blaizzy

  authored
  bbb696cView commit details

  Copy full SHA for bbb696c

  Browse repository at this point

• Add moondream3 (#807)

  Show description for 7394195
  Blaizzy

  authored
  7394195View commit details

  Copy full SHA for 7394195

  Browse repository at this point

• Adding orpo (#795)

  Show description for e025529
  

  Goekdeniz-Guelmez

  and

  Blaizzy

  authored
  e025529View commit details

  Copy full SHA for e025529

  Browse repository at this point