Skip to content

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

Commits on main

Integration test tuning (#12492)
Daniel Hiltgen committed 1 year ago
4e5d862
docs: improve accuracy of LLM library docs (#12530)
Daniel Hiltgen committed 1 year ago
303be93
win: fix build script (#12513)
Daniel Hiltgen committed 1 year ago
9182319
discovery: prevent dup OLLAMA_LIBRARY_PATH (#12514)
Daniel Hiltgen committed 1 year ago
04c1849
2c2f4de
CI: fix win arm build (#12502)
Daniel Hiltgen committed 1 year ago
292767a
CI: replace clang compiler for windows (#12495)
Daniel Hiltgen committed 1 year ago
ae5e0f0
llm: Support KV cache quantization with gpt-oss
Jesse Gross committed 1 year ago
19e6796
Fixed Deepseek2 adding nil tensor error
Grace committed 1 year ago
33801c1
Workaround broken NVIDIA iGPU free VRAM data (#12490)
Daniel Hiltgen committed 1 year ago
e434066
test: add template error test (#12489)
Patrick Devine committed 1 year ago
2fa1e92
ci: place rocm windows in correct runner dir (#12487)
Daniel Hiltgen committed 1 year ago
07e3676
CI: temporarily disable clang install (#12486)
Daniel Hiltgen committed 1 year ago
c29fb00
ci: fix windows build (#12485)
Daniel Hiltgen committed 1 year ago
730ed6e
ci: fix windows build (#12484)
Daniel Hiltgen committed 1 year ago
dc06601
1ed2881
55ca827
Update GGML to b6646 (#12245)
Daniel Hiltgen committed 1 year ago
c68f367
llm: Allow overriding flash attention setting
Jesse Gross committed 1 year ago
fdb1094
fix panic on bootstrapDevices (#12475)
Daniel Hiltgen committed 1 year ago
05a43e0
Use runners for GPU discovery (#12090)
Daniel Hiltgen committed 1 year ago
bc8909f
fix keep alive
Michael Yang committed 1 year ago
35ac4eb
ggml: Preallocate CUDA pool memory
Jesse Gross committed 1 year ago
3d0b173