Skip to content

Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

Commits on main

llm: allow iGPU mmproj offload with fit padding (#16996)
Daniel Hiltgen committed 4 months ago
a6293eb
server: remove unused internal/ code (#17071)
Patrick Devine committed 4 months ago
f3d69a3
create: harden GGUF create flows (#17062)
Daniel Hiltgen committed 4 months ago
67b6a1c
87b6421
mlx: update to de7b4ed9 (#17056)
Daniel Hiltgen committed 4 months ago
f2d069f
server: remove OLLAMA_EXPERIMENT=client2 (#16962)
Michael Yang committed 4 months ago
5208ae7
llama.cpp update (#17055)
Daniel Hiltgen committed 4 months ago
9d77957
mlx: x/create rewrite (#16919)
Patrick Devine committed 4 months ago
964ea42
llama: enable FA on CUDA CC 6.x GPUs (#16994)
Daniel Hiltgen committed 4 months ago
dba1e27
compat: use UTF-8-safe file open (#16999)
Daniel Hiltgen committed 4 months ago
e436db2
rocm: remove no longer supported devices (#17010)
Daniel Hiltgen committed 4 months ago
26acfa4
llama: clean up dead code from llama-server work (#17007)
Daniel Hiltgen committed 4 months ago
7b22ac9
agent: harness core (#16963)
Parth Sareen committed 4 months ago
a2b3a5e
docs(cloud): update retirement list (#17000)
Michael Yang committed 4 months ago
cecd265
fix cuda toolkit lookup and parallel (#16613)
Mark Ward committed 4 months ago
2ea95fb
ci: avoid unbounded parallelism (#16966)
Daniel Hiltgen committed 4 months ago
8e7be3a
mlx: tighten up gemma4 moe loading code (#16964)
Patrick Devine committed 4 months ago
710292f
launch: check for min version for hermes desktop (#16912)
Bruce MacDonald committed 4 months ago
ada1eb5
llama.cpp update (#16960)
Daniel Hiltgen committed 4 months ago
1c5ebbf
mlx: bump dependency (#16935)
Daniel Hiltgen committed 4 months ago
7926b99
d26a585
2e474c9
launch: update hermes install urls to official (#16913)
Bruce MacDonald committed 4 months ago
2cb2c53
2a6b504
CUDA: require driver 550 or newer for v12 (#16895)
Daniel Hiltgen committed 4 months ago
f22ec2e
d9075ca