Skip to content

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

Commits on main

ggml: Disable flash attention for gemma2
Jesse Gross committed 1 year ago
29ddfc2
71cb86a
docs: add ollama-co2 to community integrations (#12230)
CarbonatedWater.org committed 1 year ago
5198956
Add v12 + v13 cuda support (#12000)
Daniel Hiltgen committed 1 year ago
17a023f
8d6fffa
tests: add tool calling integration test (#12232)
Parth Sareen committed 1 year ago
20b53ea
tests: reduce stress on CPU to 2 models (#12161)
Daniel Hiltgen committed 1 year ago
6745182
f810ec7
llm: Clamp batch size to context size
Jesse Gross committed 1 year ago
e119783
runner: move harmony to runner (#12052)
Parth Sareen committed 1 year ago
1a558f9
Hybrid and recurrent memory estimates (#12186)
Gabe Goodhart committed 1 year ago
7b91c9c
docs: show how to debug nvidia init failures (#12216)
Daniel Hiltgen committed 1 year ago
950d33a
9714e38
embedding gemma model (#12181)
Michael Yang committed 1 year ago
5994e8e
more logutil.Trace (#12177)
Michael Yang committed 1 year ago
b3e6120
fb92b61
llm: Avoid underflow in free memory logging
Jesse Gross committed 1 year ago
8149a3c
harden uncaught exception registration (#12120)
Daniel Hiltgen committed 1 year ago
0cc90a8
ml: fix struct field name in comment (#12123)
pxwanglu committed 1 year ago
e42300f
66e7380
Always filter devices (#12108)
Daniel Hiltgen committed 1 year ago
ead4a9a
9d97e6a
fix keep alive (#12041)
Michael Yang committed 1 year ago
1081532
convert: fix tensor sorting (#12015)
Michael Yang committed 1 year ago
86834a2
gptoss: enable flash attention by default (#11996)
Michael Yang committed 1 year ago
85ccf73
remove extra field attr (#11205)
Michael Yang committed 1 year ago
30fb7e1