Skip to content
Arm
Git
Search or jump to…
Ctrl K
Categories
Theme: System
Sign in
Get started
armgit
/
auto-round
Public
Imported from GitHub
Fork
0
Watch
0
Star
0
PTQ-Level Efficiency, QAT-Level Accuracy | 训练后量化的效率,量化感知训练的精度
Python
diffusers
gguf
int4
llms
mxfp4
nvfp4
omni
quantization
rounding
sglang
transformers
vllm
vlms
Code
Commits
Issues
Pull requests
Wiki
Builds
Discussions
Projects
Insights
Search
main
291 branches · 45 tags
Go to file
t
Code
Commits on
main
fix mx_fp issues (#286)
wenhuach21 committed 2 years ago
99cff1f
Copy SHA
avoid deterministic algorithm warning in inference (#285)
wenhuach21 committed 2 years ago
ba5be40
Copy SHA
update readme for cpu inference
wenhuach21 committed 2 years ago
141c149
Copy SHA
update readme for v0.3.1 release (#283)
wenhuach21 committed 2 years ago
a359222
Copy SHA
refine AuoRound format and support marlin repacking (#280)
wenhuach21 committed 2 years ago
68138e8
Copy SHA
qwen2_bugfix, add adamround vision UT (#281)
WeiweiZhang1 committed 2 years ago
7cfff96
Copy SHA
refine eval (#282)
wenhuach21 committed 2 years ago
afa9e26
Copy SHA
[Important update]set full range sym as the default (#278)
wenhuach21 committed 2 years ago
00122bc
Copy SHA
adamround bugfix, refine import (#275)
WeiweiZhang1 committed 2 years ago
a633aa7
Copy SHA
change to even rounding for mantissa of mx_fp (#277)
wenhuach21 committed 2 years ago
98a9c75
Copy SHA
fix mutable default value (#272)
wenhuach21 committed 2 years ago
8bf63f3
Copy SHA
enable llama3.2-vision model quantization (#269)
WeiweiZhang1 committed 2 years ago
6b99d10
Copy SHA
keep the dtype after qdq (#268)
wenhuach21 committed 2 years ago
3a70be8
Copy SHA
remove g_idx in gptq format (#267)
wenhuach21 committed 2 years ago
fdfd971
Copy SHA
Update readme for VLM support and integration (#266)
wenhuach21 committed 2 years ago
af3db17
Copy SHA
Add a warning for improper export formats. (#265)
wenhuach21 committed 2 years ago
be32686
Copy SHA
Fix 3bit packing for auto-gptq format (#264)
wenhuach21 committed 2 years ago
6ee91a9
Copy SHA
better support quant_lm_head for larger models (#263)
wenhuach21 committed 2 years ago
82322ac
Copy SHA
refine autoawq exporting code (#261)
wenhuach21 committed 2 years ago
6539d50
Copy SHA
update eval and fix example (#260)
n1ck-guo committed 2 years ago
7816eea
Copy SHA
enable_qwen2-vl_quantization (#248)
WeiweiZhang1 committed 2 years ago
3275df9
Copy SHA
fix preci (#258)
n1ck-guo committed 2 years ago
200fcdd
Copy SHA
fix example (#256)
n1ck-guo committed 2 years ago
d8ee1c0
Copy SHA
avoid potential bug for auto-gptq 0.8 (#250)
wenhuach21 committed 2 years ago
5dd16fc
Copy SHA
support INC API (#255)
WeiweiZhang1 committed 2 years ago
11b99d2
Copy SHA
remove unused log (#252)
n1ck-guo committed 2 years ago
9b55ce5
Copy SHA
remove autoawq requirement at packing stage (#249)
n1ck-guo committed 2 years ago
98c4656
Copy SHA
enable llava int4 inference with autoround format (#237)
WeiweiZhang1 committed 2 years ago
b698213
Copy SHA
fix example bug (#245)
n1ck-guo committed 2 years ago
0d60ad2
Copy SHA
Add MANIFEST.in file to include requirements.txt (#243)
Sun, Xuehao committed 2 years ago
78c1cea
Copy SHA
Newer
Older