Skip to content

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

Commits on main

model: add bos token if configured
Michael Yang committed 1 year ago
53d2990
go.mod: Use full version for go 1.24.0
Jesse Gross committed 1 year ago
e185c08
docs: fix api examples link (#9360)
Steven Hartland committed 1 year ago
be2ac1e
server: allow vscode-file origins (#9313)
Eries Trisnadi committed 1 year ago
dc13813
runner: simplify tensor split parsing
Michael Yang committed 1 year ago
d6af13e
ml/backend/ggml: fix debug logging
Michael Yang committed 1 year ago
a59f665
Windows ARM build (#9120)
Daniel Hiltgen committed 1 year ago
688925a
d7d7e99
2db96c1
Add cuda Blackwell architecture for v12 (#9350)
Daniel Hiltgen committed 1 year ago
e12af46
llama: removed unused 'vendoring' file (#9351)
Jeffrey Morgan committed 1 year ago
3ad4bc8
0d69479
e91ae3d
docker: upgrade rocm to 6.3.3 (#8211)
José Pekkarinen committed 1 year ago
6ecd7f6
docs: rocm install link (#9346)
Chuanhui Liu committed 1 year ago
8888556
fix: add back bf16 support
Michael Yang committed 1 year ago
b16367b
4df98f3
sample: add sampling package for new engine (#8410)
Parth Sareen committed 1 year ago
0b7e167
314573b
go.mod: bump to go1.24 (#9242)
Blake Mizerany committed 1 year ago
4604b10
docs: add additional ROCm docs for building (#9066)
Jeffrey Morgan committed 1 year ago
7cfd4ae
server: group routes by category and purpose (#9270)
Blake Mizerany committed 1 year ago
68bac1e
ml: Abstract attention out of model definitions
Jesse Gross committed 1 year ago
f53f419
ml/backend/ggml: fix rms norm
Michael Yang committed 1 year ago
2192a28