Image-Text-to-Text
Transformers
Safetensors
kimi_k25
feature-extraction
compressed-tensors
conversational
custom_code
Eval Results
Instructions to use moonshotai/Kimi-K2.6 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use moonshotai/Kimi-K2.6 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="moonshotai/Kimi-K2.6", trust_remote_code=True) messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("moonshotai/Kimi-K2.6", trust_remote_code=True, dtype="auto") - Inference
- HuggingChat
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use moonshotai/Kimi-K2.6 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "moonshotai/Kimi-K2.6" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "moonshotai/Kimi-K2.6", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/moonshotai/Kimi-K2.6
- SGLang
How to use moonshotai/Kimi-K2.6 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "moonshotai/Kimi-K2.6" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "moonshotai/Kimi-K2.6", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "moonshotai/Kimi-K2.6" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "moonshotai/Kimi-K2.6", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }' - Docker Model Runner
How to use moonshotai/Kimi-K2.6 with Docker Model Runner:
docker model run hf.co/moonshotai/Kimi-K2.6
use-fast-tokenizer-v2
#40 opened 7 days ago
by
bigmoyan
Fast tokenizer token-ID parity with previous slow tokenizer?
3
#39 opened 9 days ago
by
bullpoint
Fix save_pretrained / from_pretrained round-trip for KimiK25VisionProcessor
#37 opened 9 days ago
by
fxmarty-amd
Baba Vanga test still perfectly works
🤯 1
1
#36 opened 11 days ago
by
krustik
Incorrect parameters used when calling str_replace_editor tool – view_range used with str_replace command
#35 opened 12 days ago
by
wudaolj
dev of kimi btw
#34 opened 13 days ago
by
SwifltyCW
Fix e_score_correction_bias wrong dtype
1
#33 opened 14 days ago
by
fxmarty-amd
Was Stirrup used for the 27.9% APEX-Agents-AA score?
#32 opened 14 days ago
by
pandemo
Tool-call argument JSON malformations on long-content payloads (KIMI 2.6 on VLLM 0.19.1)
#31 opened 15 days ago
by
prakharprakash
Add Claw-Eval evaluation results
#30 opened 15 days ago
by
SaylorTwift
Add YC-Bench benchmark result (avg $511,137)
#29 opened 19 days ago
by
RiddleHe
[Feature request] Eliminate pre-attention RMSNorm in MLA-models via scale invariance + weight folding
#28 opened 19 days ago
by
graefics
First melody symphony created, finally by 2 years of search (647Gb RAM)
3
#27 opened 20 days ago
by
krustik
add deepseek v4 pro to comparisons on the readme
1
#25 opened 24 days ago
by
evewashere
ored picture.
1
#24 opened 24 days ago
by
Nola21nstrom
This model introduces itself as Anthropic, if asked in Korean
12
#23 opened 26 days ago
by
picopress
OCR识别能力有待提升,qwen3.5-27b或35b-a3b,主要信息都能识别出来
1
#21 opened 27 days ago
by
htl258
arabic support ?
1
#20 opened 27 days ago
by
heisseth
Thanks for the Strong Improvements in Coding and Terminal-Bench 2.0
1
#19 opened 28 days ago
by
mfzzzzzz
Add tokenizer.json for Kimi K2.6
1
#18 opened 28 days ago
by
hoangquan456
Guide to run Kimi K2.6 on CPU, GPU and SSD setups! 🔥
👍 1
2
#17 opened 28 days ago
by
danielhanchen
Finally, this Kimi resolved my code test (Q4-X, 677Gb RAM)
#15 opened 28 days ago
by
krustik
"Modified" MIT License doesn't exist
🤯 1
2
#14 opened 29 days ago
by
jebbam
Would it be possible to take a look at how this is implemented in the source code?
4
#11 opened 30 days ago
by
Ttooong
Composer 3 is about to release very soon and it's gonna destroy this model!
😔 10
7
#9 opened 30 days ago
by
number1composerfan
thank you sir
👍 13
7
#7 opened 30 days ago
by
tom1ong
Will the zig inference code get open sourced?
2
#5 opened 30 days ago
by
ZeroAurora