Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Sep 18, 2025

• fix: model load for unsupported embedding models (#12311)

  Show description for 9f3a37f
  mxyng

  authored
  9f3a37f

  Copy full SHA for 9f3a37f

• feat: qwen3 embed (#12301)

  Show description for 7460259
  mxyng

  authored
  7460259

  Copy full SHA for 7460259

• server: add unauthorized error to remote chat handler (#12338)

  jmorganca

  authored
  22ccdd7

  Copy full SHA for 22ccdd7

• build: avoid unbounded parallel builds (#12319)

  Show description for 0c3d0e7
  dhiltgen

  authored
  0c3d0e7

  Copy full SHA for 0c3d0e7

• auth: check the permissions on the private key to see if it's readable (#12336)

  pdevine

  authored
  eb0a5d4

  Copy full SHA for eb0a5d4

• fix(integration): check truncated length (#12337)

  mxyng

  authored
  ceac416

  Copy full SHA for ceac416

• convert: convert bf16 vision weights to fp16 (#12324)

  Show description for 2717dce
  pdevine

  authored
  2717dce

  Copy full SHA for 2717dce

Commits on Sep 17, 2025

• server: skip parsing initial <think> if provided in the prompt for /api/generate (#12289)

  rick-github

  authored
  9b8187b

  Copy full SHA for 9b8187b

• engine: add remote proxy (#12307)

  pdevine

  authored
  8b89493

  Copy full SHA for 8b89493

• fix: multi-cuda version skew (#12318)

  Show description for 9c5bf34
  dhiltgen

  authored
  9c5bf34

  Copy full SHA for 9c5bf34

• fix(llama): other llama flavours (#12308)

  Show description for 564b558
  mxyng

  authored
  564b558

  Copy full SHA for 564b558

• prefer ollama engine for qwen3 (#12310)

  mxyng

  authored
  a417ac9

  Copy full SHA for a417ac9

• refactor: use the built-in max/min to simplify the code (#12280)

  Show description for 05d5345
  russcoss

  authored
  05d5345

  Copy full SHA for 05d5345

Commits on Sep 16, 2025

• logutil: fix source field (#12279)

  mxyng

  authored
  b225508

  Copy full SHA for b225508

• Merge pull request #12248 from ollama/drifkin/qwen3-coder-parsing

  Show description for fa1c987
  drifkin

  authored
  fa1c987

  Copy full SHA for fa1c987

• use split activations when possible (#12293)

  Show description for ad95d5b
  mxyng

  authored
  ad95d5b

  Copy full SHA for ad95d5b

• embed: cleanup (#12299)

  Show description for c253433
  mxyng

  authored
  c253433

  Copy full SHA for c253433

• fix: fix CUDA detection for older GPUs (#12300)

  Show description for a1cff89
  egyptianbman

  authored
  a1cff89

  Copy full SHA for a1cff89

Commits on Sep 15, 2025

• doc: show how to clear the cgo cache (#12298)

  dhiltgen

  authored
  93c64ea

  Copy full SHA for 93c64ea

• model: implement bert in ollama engine (#9080)

  Show description for 3f6642f
  mxyng

  authored
  3f6642f

  Copy full SHA for 3f6642f

• batch: use tensors for outputs (#12185)

  Show description for 6f71171
  mxyng

  authored
  6f71171

  Copy full SHA for 6f71171

• address comments

  drifkin

  committed
  472feec

  Copy full SHA for 472feec

• add qwen3-coder tool support

  Show description for 4799194
  drifkin

  committed
  4799194

  Copy full SHA for 4799194

Commits on Sep 12, 2025

• Revert "runner: move harmony to runner (#12052)"

  Show description for 92b96d5
  jmorganca

  committed
  92b96d5

  Copy full SHA for 92b96d5

• Revert "runner: simplify parser entrypoints in runner (#12233)"

  Show description for 9d56e63
  jmorganca

  committed
  9d56e63

  Copy full SHA for 9d56e63

• Fix image cannot be seen with slice image on llama engine

  Show description for 0530921
  tc-mb

  authored
  0530921

  Copy full SHA for 0530921

• tests: tighten up a few flaky tests (#12271)

  Show description for 44a6792
  dhiltgen

  authored
  44a6792

  Copy full SHA for 44a6792

• cuda: remove compression for better compatibility (#12259)

  Show description for e4ce683
  dhiltgen

  authored
  e4ce683

  Copy full SHA for e4ce683

Commits on Sep 11, 2025

• ollamarunner: Suppress stack trace during memory allocation

  Show description for 2621412
  jessegross

  committed
  2621412

  Copy full SHA for 2621412

• CI: fix windows cuda build (#12246)

  Show description for 61fb912
  dhiltgen

  authored
  61fb912

  Copy full SHA for 61fb912

• llm: Don't try to load split vision models in the Ollama engine

  Show description for aba1575
  jessegross

  committed
  aba1575

  Copy full SHA for aba1575

• llm: Enable new memory estimates by default

  Show description for eb10390
  jessegross

  committed
  eb10390

  Copy full SHA for eb10390

• feat: add dimensions field to embed requests (#12242)

  Show description for feb18cd
  mxyng

  authored
  feb18cd

  Copy full SHA for feb18cd

• cmd: use slices.Contains to simplify code (#12249)

  fengyuchuanshen

  authored
  8a7e205

  Copy full SHA for 8a7e205

Commits on Sep 10, 2025

• ggml: Disable flash attention for gemma2

  Show description for 29ddfc2
  jessegross

  committed
  29ddfc2

  Copy full SHA for 29ddfc2

Pagination

Previous

Next