Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
Commits on main
qwen3next: avoid inplace sigmoid for shared gate (#14077)
d25535c Jeffrey Morgan committed 9 months ago
cmd: helpful error message for remote models (#14057)
c323161 Bruce MacDonald committed 9 months ago
runner: discard compute results if sequence replaced mid-batch (#14072)
f7102ba Jeffrey Morgan committed 9 months ago
Revert "cmd: claude launch improvements (#14064)" (#14071)
cefabd7 Jeffrey Morgan committed 9 months ago
server: optimize chatPrompt to reduce tokenization calls (#14040)
df70249 Jeffrey Morgan committed 9 months ago
cmd: open browser on `ollama signin` when available (#14055)
a635532 Bruce MacDonald committed 9 months ago
anthropic: add InputTokens to streaming response (#13934)
7189648 Aleksandr Vukmirovich committed 9 months ago
llm: Make "do load request" error message more informative
6582f6d Richard Lyons committed 9 months ago
server: fix ollama ps showing configured instead of actual context length
d11fbd2 Jesse Gross committed 9 months ago
runner: fix typo 'baackend' -> 'backend' in error messages (#13645)
3590fbf Dhiraj Lochib committed 9 months ago
docs: add Screenpipe to Community Integrations (#13906)
06bc8e6 Louis Beaumont committed 9 months ago
cmd: ollama launch add placeholder text for selector (#13966)
a0923cb Parth Sareen committed 9 months ago