Skip to content

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

Commits on main

108fe02
conditionally enable parallel pipelines
Michael Yang committed 1 year ago
4561fff
Add support for ROCm gfx1151 (#9773)
Daniel Hiltgen committed 1 year ago
50b5962
e27e4a3
fix: correctly save in interactive mode (#9788)
Patrick Devine committed 1 year ago
2c8b484
gemma3 quantization (#9776)
Patrick Devine committed 1 year ago
ef378ad
Align versions for local builds (#9635)
Daniel Hiltgen committed 1 year ago
2d2247e
gemma3: Allow multiple image in a single input
Jesse Gross committed 1 year ago
7bf793a
282bfaa
9679f40
3892c3a
server/internal/chunks: remove chunks package (#9755)
Blake Mizerany committed 1 year ago
4e320b8
4ea4d2b
count non-repeating vision layers
Michael Yang committed 1 year ago
8d76fa2
74b44fd
fix divide by zero
Michael Yang committed 1 year ago
65b88c5
roughly count gemma3 graph
Michael Yang committed 1 year ago
a422ba3
count all vision tensors
Michael Yang committed 1 year ago
d2ec223
count gemma3 vision tensors
Michael Yang committed 1 year ago
033cec2
543240f
add verbose mode to the show command (#9640)
Patrick Devine committed 1 year ago
4bed739
fix: change default context size for gemma3 (#9744)
Patrick Devine committed 1 year ago
80c7ce3
Update model/model.go
Michael Yang committed 1 year ago
3e102b7
ec46f32
fix: error if image requested without vision model
Michael Yang committed 1 year ago
5e2e0b4