moonshotai/Kimi-K2.5 · Discussions

Add MathArena evaluation result for hmmt/hmmt_feb_2026

#95 opened about 24 hours ago by

JasperDekoninck

Very good model for simulating Baba Vanga

1

#94 opened 1 day ago by

krustik

About Kimi Code

#93 opened 2 days ago by

llMss12

clone this

1

#92 opened 4 days ago by

givikjathan

fix AttributeError: 'MoonViT3dEncoder' object has no attribute 'use_deterministic_attn'

#91 opened 5 days ago by

katuni4ka

Add Terminal-Bench evaluation result (43.2%)

#90 opened 5 days ago by

burtenshaw

Possibility of a 10b model?

#89 opened 6 days ago by

crownelius

Add MathArena evaluation result for aime/aime_2026

#88 opened 6 days ago by

JasperDekoninck

Does Kimi2.5 only score 72.57 on tau2-bench?

#87 opened 10 days ago by

Alicia-Ross

All good

2

#86 opened 11 days ago by

suprapto089

pathed_the_params.bash

#80 opened 12 days ago by

Kjppmp

Can someone abliterate this model please

2

#79 opened 13 days ago by

kljnljnj

BF16/FP8 version availability?

2

#77 opened 13 days ago by

JFLOR274

Ai _chAt bot

1

#74 opened 14 days ago by

vijaybarman3

The wave

#73 opened 14 days ago by

Nk98James

Kimi K2.5 slow?

2

#72 opened 14 days ago by

sebastienbo

Knowledge Cut-off

#71 opened 15 days ago by

virtualhat

H200 - zero gpu

#70 opened 15 days ago by

rahul7star

未正确识别工具调用

2

#69 opened 15 days ago by

yeguangsuixing

age

➕ 1

1

#68 opened 16 days ago by

DannySc

Model parameters confusion

1

#67 opened 17 days ago by

Ujjwal-Tyagi

Add converted tokenizer (no trust_remote_code needed)

2

#66 opened 18 days ago by

ArthurZ

first commit

#65 opened 19 days ago by

pika19

success

#64 opened 19 days ago by

pika19

多图输入的占位符或顺序？

4

#63 opened 19 days ago by

noobimp

Value error, Model architectures ['KimiK25ForConditionalGeneration'] are not supported for now. Supported architectures:

1

#62 opened 19 days ago by

jianyoulin

performance of quantized models

3

#61 opened 19 days ago by

darvec

400 Bad Request when deploying Kimi-K2.5 with vllm-0.15.0

#60 opened 19 days ago by

tuo02

hi, who are you? -> I am Claude.

6

#59 opened 19 days ago by

Jeol

Model

1

#57 opened 20 days ago by

Vickweb3

Navigation of hugging face

#55 opened 20 days ago by

Xaviercrypt01

Model not generating </think> token .

➕ 6

3

#52 opened 20 days ago by

pratiknarola

Add MMLU-Pro evaluation results

#51 opened 20 days ago by

nielsr

Add GPQA diamond evaluation results

#50 opened 20 days ago by

nielsr

Add HLE evaluation results (with tools)

🚀 1

#49 opened 20 days ago by

nielsr

The content of response from Kimi-K2.5 is empty when deploying with vllm-0.15.0.

5

#40 opened 21 days ago by

tuo02

AttributeError: 'CompressedLinear' object has no attribute 'weight'

#39 opened 21 days ago by

xbeark

Kimi K2.5 claims to be Claude, behaves like it, and even references Anthropic's Constitutional AI.

🤗 👀 12

5

#38 opened 22 days ago by

ombd

Update chat template to use reasoning if reasoning_content is unavailable, which is deprecated

1

#37 opened 22 days ago by

koushd

fix: wrong return type annotation in resampling() function

#36 opened 22 days ago by

ThanhNguyxn

fix: type annotation bugs in media_utils, processor and config

#35 opened 24 days ago by

ThanhNguyxn

fix: bugs and improvements

1

#34 opened 24 days ago by

ThanhNguyxn

Can it be added to microsoft foundry please?

1

#32 opened 25 days ago by

atootikhos4

返回的内容为claude的思考过程的数据格式

🤗 👍 3

2

#30 opened 25 days ago by

tzzhangyajun

Kimi K2.5 using ktkernel + sglang, 16 TPS, but no starting <think> tag.

➕ 1

5

#28 opened 26 days ago by

gyularabai

Where is the mecord?

2

#25 opened 26 days ago by

luoxlong

Request: DOI

#23 opened 27 days ago by

torouni

bug: tools with dashes in their name

➕ 2

2

#22 opened 27 days ago by

koushd

Need speed estimate on 6x H200

2

#21 opened 27 days ago by

mechanicmuthu

Guide to run Kimi K2.5 locally on your device.

😔 🤯 12

2

#19 opened 27 days ago by

shimmyshimmer