Token Count Calculation in SFT Data Distribution Curation
2
#31 opened 2 days ago
by
tcy006
kimi-k2-thinking can't use a tool with 2 required arguments if it has 2 or more tools.
#30 opened 4 days ago
by
Yerkhat
Is the reasoning parser going to be available in vllm soon?
1
#29 opened 5 days ago
by
dingobingobango
If we apply PTQ to a QAT model, what will happen
#28 opened 6 days ago
by
onekq
User Reviews about K2 Thinking and Search Agent
1
#27 opened 6 days ago
by
DeepNLP
Was this really trained during QAT using a symmetric 4bit quant with only 15/16 values used?
π
9
3
#26 opened 6 days ago
by
jukofyork
Their API Takes Your Data - PRIVACY RISK
9
#24 opened 7 days ago
by
evilperson068
AssertionError when hosting via vLLM in H20x8
2
#23 opened 7 days ago
by
O-delicious
Any plan to open source the search agent framework?
β
3
6
#22 opened 7 days ago
by
CherryDurian
Beware quantization destroyed coding abilities, broken code in Q5-K-M (735Gb RAM)
9
#21 opened 7 days ago
by
krustik
Does the Ktranformers deployment support non-AVX-512?
1
#16 opened 10 days ago
by
bullerwins
Question: Will models with other precisions be published soon?
π
1
1
#14 opened 10 days ago
by
jhv00
Question: What model precision was used during the benchmarks?
#13 opened 10 days ago
by
evewashere
vllm deployment failed
8
#11 opened 10 days ago
by
mondaylord
How to control reasoning effort? Heavy Mode/low-latency mode?
#10 opened 10 days ago
by
huggingfacemotnt
Did you set GPT-5's reasoning effort to high?
2
#6 opened 11 days ago
by
madmax0404
K2 Thinking Browsecomp/HLE Reproducibility | η»ζε€η°
β
2
17
#5 opened 11 days ago
by
pandemo
Awesome work! Do you want to try AMO-Bench, the most challenging MO-level benchmark?
#3 opened 11 days ago
by
ShengnanAn
Python script to decompress tensors?
β
π₯
2
11
#2 opened 12 days ago
by
ubergarm
Video of Step-by-Step Review and Testing
π
π
6
#1 opened 12 days ago
by
fahdmirzac