Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
Commits on main
integration: improve ability to test individual models (#14948)
c9b5da6 Daniel Hiltgen committed 8 months ago
mlxrunner: support partial match on pure transformer caches
7749143 Jesse Gross committed 8 months ago
launch: skip openclaw gateway health check when no daemon install (#14984)
6df6d09 Bruce MacDonald committed 8 months ago
llm, mlxrunner: fix done channel value consumed by first receiver
d7c176a Jesse Gross committed 8 months ago
app: fix desktop app stuck loading when OLLAMA_HOST is an unspecified bind address (#14885)
3f3a24b Eva H committed 8 months ago
mlxrunner: share KV cache across conversations with common prefixes
96e36c0 Jesse Gross committed 8 months ago
mlxrunner: fix Slice(0, 0) returning full dimension instead of empty
6f8ddbb Jesse Gross committed 8 months ago
cmd/launch: skip redundant config writes when model unchanged (#14941)
b5e7888 Eva H committed 8 months ago
docs: update claude code and openclaw for web search (#14922)
eab4d22 Parth Sareen committed 8 months ago
launch: fix openclaw not picking up newly selected model (#14943)
5759c2d Bruce MacDonald committed 8 months ago
docs: update minimax-m2.5 references to m2.7 (#14942)
42b1c26 Bruce MacDonald committed 8 months ago
launch: disable claude attribution header to preserve KV cache
f622b0c Jesse Gross committed 8 months ago
cmd/launch: check for both npm and git before installing OpenClaw (#14888)
5d00006 Bruce MacDonald committed 8 months ago
cloud_proxy: for the web_search legacy path, flush on newlines (#14897)
e37a9b4 Devon Rifkin committed 8 months ago
mlx: quantized embeddings, fast SwiGLU, and runtime fixes (#14884)
d727aac Patrick Devine committed 8 months ago
mlx: add prequantized tensor packing + changes for qwen35 (#14878)
fa69b83 Patrick Devine committed 8 months ago