-
Notifications
You must be signed in to change notification settings - Fork 17.9k
Pull requests: ollama/ollama
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
server: separate generate thinking from non-thinking parser output
#18561
opened Sep 20, 2026 by
lorenzozanee
Loading…
mlxrunner: make prefix-cache paged-out memory budget configurable
#18556
opened Sep 20, 2026 by
nnysted
Loading…
handle NaN totalRawPower when SMCount/ClockMHz unavailable
#18552
opened Sep 20, 2026 by
c4801725870
Loading…
Warn and ignore typical_p on API requests instead of rejecting them
#18551
opened Sep 20, 2026 by
SIDDARTHAREDDY8
Loading…
openai: accept reasoning_content in assistant messages
#18543
opened Sep 19, 2026 by
isharak7m
Loading…
model/parsers: recognize <function= as an implicit qwen3-coder tool-call opener
#18538
opened Sep 19, 2026 by
farmerstevieb
Loading…
model/parsers: accept implicit qwen3coder tool openers
#18532
opened Sep 18, 2026 by
breken-ai
Loading…
Add digest and provider_version to /api/chat responses
#18524
opened Sep 18, 2026 by
xiaoyulox
Loading…
Allow 96-char model names so long HF GGUF pulls validate
#18520
opened Sep 18, 2026 by
ege-arhan
Loading…
server: serve Prometheus metrics at /metrics when OLLAMA_METRICS is set
#18508
opened Sep 17, 2026 by
ssubbotin
Loading…
fix(server): parse reasoning on /api/generate even when think is false
#18504
opened Sep 17, 2026 by
AloysJehwin
Loading…
fix(gemma4): stop dropping tool parameters named like schema keywords
#18503
opened Sep 17, 2026 by
AloysJehwin
Loading…
Fix: minicpm5-2b native tool calls never parse
#18499
opened Sep 17, 2026 by
SIDDARTHAREDDY8
Loading…
Previous Next
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.