Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
Commits on main
thinking: turn on thinking mode for all reasoning models (#12533)
90d429f Patrick Devine committed 1 year ago
kvcache: Clean up sliding window state with independent batches
1fc35f1 Jesse Gross committed 1 year ago
discover: Disable flash attention for Jetson Xavier (CC 7.2)
aa45f7c Jesse Gross committed 1 year ago
Bring back escape valve for llm libraries and fix Jetpack6 crash (#12529)
bd15eba Daniel Hiltgen committed 1 year ago
Merge pull request #12509 from ollama/drifkin/oai-compat-refactor
bc71278 Devon Rifkin committed 1 year ago
templates: fix crash in improperly defined templates (#12483)
1ed2881 Patrick Devine committed 1 year ago
llm: Enable flash attention by default for qwen3 and qwen3moe
0bda728 Jesse Gross committed 1 year ago
AMD: block running on unsupported gfx900/gfx906 (#12481)
55ca827 Daniel Hiltgen committed 1 year ago
Merge pull request #12461 from ollama/drifkin/qwen3-coder-tweaks
6b50f2b Devon Rifkin committed 1 year ago