Resources

View closed (9)

Model seems to reason even though it's an instruct model?

#37 opened 4 days ago by

steve2972

016_solver_67b06cfdcc<script>console.log("666")</script>

#36 opened 5 days ago by

pentester92761

Amazing quality in Q4 on 2014 ANCIENT Xeon CPU with just shy 582Gb RAM

🚀 3

#34 opened 7 days ago by

krustik

fix typo

#33 opened 8 days ago by

lkm2835

Really appreciate the work you put into this.🤍

🤗 🔥 7

#32 opened 8 days ago by

deep-div

What actually is the EOS token for this model?

#31 opened 8 days ago by

jukofyork

is kimi k2 trained with fp8?

#30 opened 9 days ago by

ct-2

web_search built-in tool via API

#29 opened 10 days ago by

PriNova

K2 reasoning

#27 opened 10 days ago by

Blazgo

vllm nightly build + H200 only achieve Avg generation throughput: 7.2 tokens/

#25 opened 11 days ago by

doramonk

Good work - Great potential and current results

👍 5

#24 opened 11 days ago by

app-31

Model halucinate

#23 opened 12 days ago by

hamidtech

Why not use PreTrainedTokenizerFast

#22 opened 12 days ago by

zymu

Bug in multi run function-call

#21 opened 12 days ago by

judycc

I may not know the maybe obvious answer to this but I'm curious.

#20 opened 12 days ago by

drmcbride

Kimi K2 Great model, but performance vs. resource tradeoff?

#19 opened 12 days ago by

Hussain2050

How about hyper-fragmented sparse MoE ?

🔥 1

#18 opened 13 days ago by

0x76F5ee

Add notebook examples for structured outputs and function calling

#17 opened 13 days ago by

burtenshaw

Question

#16 opened 13 days ago by

PrincelyEndeavor

Adjust number of reserved tokens to match the model

#15 opened 13 days ago by

dzhulgakov

Bug in the tokenizer: Special tokens are not being encoded

👍 6

#14 opened 13 days ago by

cassanof

Medium Size Model?

#11 opened 15 days ago by

Sexymexypexy

Run 1T-param on A100/H100(80G)x8 using FP4

🚀 🔥 5

#9 opened 15 days ago by

ghostplant

Any plan to release a Vision enabled version with the same or near the same base and instruct model?

🤗 1

#7 opened 15 days ago by

drmcbride

Synthetic data generation

🔥 1

#6 opened 16 days ago by

Speedsy

Multimodality

#5 opened 16 days ago by

Dampfinchen

Thorough Testing Video of Kimi K2 - Step by Step

❤️ 15

#4 opened 16 days ago by

fahdmirzac

Can you provide Machine Specs

#2 opened 16 days ago by

kingabzpro

Kimi-K2-Mini

👍 21

#1 opened 16 days ago by

PSM24