ollama

server: hide duplicate and downgrade guards from list (#18874)

release.yaml #148 -Commit 531946e996 pushed by CERBERUS

v0.40.2-rc0

2026-10-10 00:43:02 +00:00

0s

Merge pull request #18855 from ollama/launch-claude-max-context-tokens: Set CLAUDE_CODE_MAX_CONTEXT_TOKENS to model's context length

release.yaml #147 -Commit b061384d90 pushed by CERBERUS

v0.40.2

2026-10-10 00:43:01 +00:00

0s

mlx: drop carried metal residency patch now that it is upstream (#18854)

release.yaml #146 -Commit cf2a313a29 pushed by CERBERUS

v0.40.1-rc0

2026-10-09 06:49:41 +00:00

0s

mlx: drop carried metal residency patch now that it is upstream (#18854)

release.yaml #145 -Commit cf2a313a29 pushed by CERBERUS

v0.40.1

2026-10-09 06:49:38 +00:00

0s

routes: fix embed 413 vs 400 routing (#18822)

release.yaml #144 -Commit 0d0720e51f pushed by CERBERUS

v0.40.0

2026-10-08 06:03:49 +00:00

0s

model: add multimodal embeddings (#18820)

release.yaml #143 -Commit efe43c5561 pushed by CERBERUS

v0.40.0-rc6

2026-10-07 18:26:52 +00:00

0s

mlx: fix patch for recent mlx update (#18812)

release.yaml #142 -Commit 8a971df3ba pushed by CERBERUS

v0.40.0-rc5

2026-10-07 12:38:20 +00:00

0s

MLX: version bump (#18720)

release.yaml #141 -Commit 18846f8a9b pushed by CERBERUS

v0.40.0-rc4

2026-10-07 12:38:19 +00:00

0s

pull: allow RCs to pull matching min_version (#18790)

release.yaml #140 -Commit 991421d4db pushed by CERBERUS

v0.40.0-rc3

2026-10-06 06:10:52 +00:00

0s

Merge remote-tracking branch 'upstream/main' into release_v0.40.0

release.yaml #139 -Commit 0fde06c4d2 pushed by CERBERUS

v0.40.0-rc2

2026-10-06 00:28:48 +00:00

0s

mlx: match publisher tokenizer semantics (#18779)

release.yaml #138 -Commit 195f4cdca1 pushed by CERBERUS

v0.40.0-rc1

2026-10-06 00:26:06 +00:00

0s

ci: fix missing build context (#18742)

release.yaml #137 -Commit b0c1ca4f75 pushed by CERBERUS

v0.35.1

2026-10-03 18:36:51 +00:00

0s

ci: fix missing build context (#18742)

release.yaml #136 -Commit b0c1ca4f75 pushed by CERBERUS

v0.35.1-rc2

2026-10-03 12:31:23 +00:00

0s

models: add clef support (#18741)

release.yaml #135 -Commit e4c0d18eb8 pushed by CERBERUS

v0.35.1-rc1

2026-10-03 12:20:57 +00:00

0s

create: support explicit model capabilities (#18708)

release.yaml #134 -Commit b3f78b7732 pushed by CERBERUS

v0.35.1-rc0

2026-10-01 00:22:36 +00:00

0s

mlx: bound pull stall retries and let the watchdog interrupt them (#18625)

release.yaml #133 -Commit cc4069396f pushed by CERBERUS

v0.35.0-rc1

2026-09-30 06:04:10 +00:00

0s

feat: add System One scoring API (#18606)

release.yaml #132 -Commit ffb6f220d0 pushed by CERBERUS

v0.35.0-rc0

2026-09-30 06:04:09 +00:00

0s

mlx: bound pull stall retries and let the watchdog interrupt them (#18625)

release.yaml #131 -Commit cc4069396f pushed by CERBERUS

v0.35.0

2026-09-30 06:04:08 +00:00

0s

llama-server: prepare to remove compatibility patch

release.yaml #130 -Commit 75b952780f pushed by CERBERUS

v0.40.0-rc0

2026-09-26 12:03:54 +00:00

0s

mlxrunner: Update XGrammar to 0.2.7 for structured outputs

release.yaml #129 -Commit b2da9e468a pushed by CERBERUS

v0.34.4-rc1

2026-09-25 12:04:10 +00:00

0s

mlxrunner: Update XGrammar to 0.2.7 for structured outputs

release.yaml #128 -Commit b2da9e468a pushed by CERBERUS

v0.34.4

2026-09-25 12:04:09 +00:00

0s

mlx: speed up Qwen 3.8 prompt processing (#18550)

release.yaml #127 -Commit c0f8da35b2 pushed by CERBERUS

v0.34.4-rc0

2026-09-24 12:04:01 +00:00

0s

server: allow registry cross-host redirects among allowlisted hosts (#18533)

release.yaml #126 -Commit 6383a0fa9c pushed by CERBERUS

v0.34.3

2026-09-24 00:04:12 +00:00

0s

server: allow registry cross-host redirects among allowlisted hosts (#18533)

release.yaml #125 -Commit 6383a0fa9c pushed by CERBERUS

v0.34.3-rc1

2026-09-20 03:14:48 +00:00

0s

api: expose model thinking levels and defaults (#18473)

release.yaml #124 -Commit d0c8cdb795 pushed by CERBERUS

v0.34.3-rc0

2026-09-20 03:13:04 +00:00

0s

x/transfer, server: tighten redirect handling for registry requests (#18512)

release.yaml #123 -Commit dfabde4539 pushed by CERBERUS

v0.34.2

2026-09-19 09:17:00 +00:00

0s

cli: add first-run onboarding shared with the desktop app (#18495)

release.yaml #122 -Commit 7fa8a024b7 pushed by CERBERUS

v0.34.2-rc3

2026-09-18 21:13:05 +00:00

0s

mlxrunner: Release freed KV buffers during speculative decode

release.yaml #121 -Commit ec3cc23076 pushed by CERBERUS

v0.34.2-rc2

2026-09-18 21:13:04 +00:00

0s

mlxrunner: lay out model by contract, checkpoint and construction

release.yaml #120 -Commit 66c3238b3d pushed by CERBERUS

v0.34.2-rc1

2026-09-18 05:28:47 +00:00

0s

llama.cpp: version bump b10969 (#18446)

release.yaml #119 -Commit a43fad18b0 pushed by CERBERUS

v0.34.2-rc0

2026-09-17 03:07:33 +00:00

0s