Resources

View closed (13)

Add GPQA evaluation result

#52 opened 7 days ago by

burtenshaw

I think `eos_token_id` should be `163585`in `config.json`?

#51 opened 7 days ago by

jukofyork

Add community evaluation results for GPQA

#50 opened 19 days ago by

nielsr

Add MMLU-Pro evaluation result

#49 opened 19 days ago by

nielsr

Add HLE evaluation result

#48 opened 20 days ago by

burtenshaw

New tool to benchmark GGUF files

#47 opened about 1 month ago by

Nerdsking

This model thinks that repeating itself is good

👀 1

#46 opened about 2 months ago by

ChuckMcSneed

Which Longform Writing benchmark is?

#44 opened about 2 months ago by

woolpeeker

Update tokenization_kimi.py

🔥 2

#43 opened about 2 months ago by

noamw-aai

need the version of transformer

#42 opened about 2 months ago by

Jack879

Kimi-K2-Thinking vLLM self host tool call fail

#40 opened about 2 months ago by

CHONGYOEYAT

I need a teacher

#39 opened about 2 months ago by

0xola

How to limit the thinking duration?

#38 opened about 2 months ago by

Spankyboy

Add Artificial Analysis evaluations for kimi-k2-thinking

#37 opened 2 months ago by

mackenzietechdocs

tokenizer.apply_chat_template is right? Manually Parsing Tool Calls Failed! but openai api Success, Why?

#36 opened 3 months ago by

nwzjk

Kimi-K2 (non-thinking) QAT

#35 opened 3 months ago by

geboh67859

Token Count Calculation in SFT Data Distribution Curation

#31 opened 3 months ago by

tcy006

kimi-k2-thinking can't use a tool with 2 required arguments if it has 2 or more tools.

#30 opened 3 months ago by

Yerkhat

Is the reasoning parser going to be available in vllm soon?

#29 opened 3 months ago by

dingobingobango

If we apply PTQ to a QAT model, what will happen

#28 opened 3 months ago by

onekq

User Reviews about K2 Thinking and Search Agent

#27 opened 3 months ago by

DeepNLP

Was this really trained during QAT using a symmetric 4bit quant with only 15/16 values used?

👀 9

#26 opened 3 months ago by

jukofyork

Their API Takes Your Data - PRIVACY RISK

#24 opened 3 months ago by

evilperson068

Any plan to open source the search agent framework?

➕ 3

#22 opened 3 months ago by

CherryDurian

Beware quantization destroyed coding abilities, broken code in Q5-K-M (735Gb RAM)

#21 opened 3 months ago by

krustik

Does the Ktranformers deployment support non-AVX-512?

#16 opened 3 months ago by

bullerwins

Question: Will models with other precisions be published soon?

👍 1

#14 opened 3 months ago by

jhv00

Question: What model precision was used during the benchmarks?

#13 opened 3 months ago by

evewashere

vllm deployment failed

#11 opened 3 months ago by

mondaylord

How to control reasoning effort? Heavy Mode/low-latency mode?

👍 5

#10 opened 3 months ago by

huggingfacemotnt

MoEQuant?

#7 opened 3 months ago by

ekurtic

Did you set GPT-5's reasoning effort to high?

#6 opened 3 months ago by

madmax0404

K2 Thinking Browsecomp/HLE Reproducibility | 结果复现

➕ 2

#5 opened 3 months ago by

pandemo

Awesome work! Do you want to try AMO-Bench, the most challenging MO-level benchmark?

#3 opened 3 months ago by

ShengnanAn

Python script to decompress tensors?

🔥 ➕ 3

#2 opened 3 months ago by

ubergarm

Video of Step-by-Step Review and Testing

😎 🚀 6

#1 opened 3 months ago by

fahdmirzac