[Bug] RoPE initialization for SWA layers modifies shared config object, causing incorrect rope_theta for non-SWA layers
#32 opened 6 days ago
by
sirusGray
Add HLE evaluation result (22.1)
#31 opened 10 days ago
by
burtenshaw
Will supporting mtp model in modeling？
#27 opened 11 days ago
by
sirusGray
Infinite CoT (Chain of Thought) Loop / Failure to converge Claude Code
#26 opened 16 days ago
by
Svyatoblood
llama-factory sft 微调
#25 opened 24 days ago
by
liuweixiong
Add pipeline_tag and link to technical report
#24 opened about 1 month ago
by
nielsr
Testing over the last days
2
#23 opened about 1 month ago
by
PriNova
Nice gpt-oss distill
❤️
2
2
#22 opened about 1 month ago
by
ChuckMcSneed
Can this be loaded in transformers? FP8Linear: size mismatch for weight_scale_inv
#21 opened about 1 month ago
by
mratsim
Clarification on default prompt template
#20 opened about 1 month ago
by
dont-care
Great Model! - sglang mtp support for triton backend
👍
3
4
#19 opened about 2 months ago
by
chriswritescode
Working Config for H200 Here
2
#18 opened about 2 months ago
by
JeffersonNunn
Q8 and abliteration
👀
1
#17 opened about 2 months ago
by
testai111
Browsecomp Reproducibility | 结果复现
👍
2
#16 opened about 2 months ago
by
pandemo
工具调用功能调用第一次成功后，之后回答都是乱码
1
#15 opened about 2 months ago
by
YUN324
flashinfer报错， 是否支持 RTX 6000pro blackwell？
4
#14 opened about 2 months ago
by
lluu8
作为langchain-agent基座存在工具调用不稳定的问题
3
#12 opened about 2 months ago
by
BoChen2333
这是自研的还是微调，蒸馏的？？
👍
🧠
5
9
#8 opened about 2 months ago
by
ChangXingyang
Video of Step-by-Step Review and Testing
❤️
2
#7 opened about 2 months ago
by
fahdmirzac
Web-site Xiaomi Mimo Studio doesn't open.
3
#5 opened about 2 months ago
by
NikolayKozloff
请问会有多模态的版本吗？
#4 opened about 2 months ago
by
thesby
初测这模型可以进开源模型第一梯队
13
#2 opened about 2 months ago
by
pypry