Skip to content

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

Commits on main

fix vision encoder
Michael Yang committed 1 year ago
f888912
ollama debug tensor
Michael Yang committed 1 year ago
9e4642e
duplicate token_embd to output
Michael Yang committed 1 year ago
6b0486c
skip repacking vision tensors
Michael Yang committed 1 year ago
d368c03
fix configs
Patrick Devine committed 1 year ago
9b54267
update model
Michael Yang committed 1 year ago
46bb016
use fast attention
Michael Yang committed 1 year ago
8934324
Fix tests and drift from main
Jesse Gross committed 1 year ago
0e88659
fix conversion
Patrick Devine committed 1 year ago
c62861f
set non-causal attention
Michael Yang committed 1 year ago
0df1800
temporary work around for converting spm
Patrick Devine committed 1 year ago
631fecc
fix drift from main
Jesse Gross committed 1 year ago
4346c24
add gemma vision encoder
Michael Yang committed 1 year ago
4b037a9
gemma2 impl
Patrick Devine committed 1 year ago
5f74d1f
Build release for windows with local script (#9636)
Daniel Hiltgen committed 1 year ago
4dcf801
Merge pull request #9590 from ollama/mxyng/dump-pad
Michael Yang committed 1 year ago
26a2699
fix: pad tensor item if ge zero
Michael Yang committed 1 year ago
9926eae
8585b7b
Merge pull request #9569 from dwt/patch-1
Michael Yang committed 1 year ago
fe77629
d8a5d96
docs: add SwiftChat (#9540)
Xiaowei Zhu committed 1 year ago
757668c
docs(tool): add mcp-llm (#9537)
Sam committed 1 year ago
96ec8af
ml: Add support for quantized KV cache
Jesse Gross committed 1 year ago
4100ed7
kvcache: Set context for shift offsets
Jesse Gross committed 1 year ago
f52b261
25f9b15
kvcache: Support non-causal attention
Jesse Gross committed 1 year ago
6da8b6a