Image-Text-to-Text
Transformers
kimi_k3
feature-extraction
compressed-tensors
conversational
custom_code
Instructions to use PYTHAI/Kimi-K3-fork with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use PYTHAI/Kimi-K3-fork with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="PYTHAI/Kimi-K3-fork", 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("PYTHAI/Kimi-K3-fork", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use PYTHAI/Kimi-K3-fork with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "PYTHAI/Kimi-K3-fork" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "PYTHAI/Kimi-K3-fork", "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/PYTHAI/Kimi-K3-fork
- SGLang
How to use PYTHAI/Kimi-K3-fork 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 "PYTHAI/Kimi-K3-fork" \ --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": "PYTHAI/Kimi-K3-fork", "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 "PYTHAI/Kimi-K3-fork" \ --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": "PYTHAI/Kimi-K3-fork", "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 PYTHAI/Kimi-K3-fork with Docker Model Runner:
docker model run hf.co/PYTHAI/Kimi-K3-fork
Download FORK.json from PYTHAI/Kimi-K3-fork: direct link, hf CLI and curl.
- Browser
- Download file 3.11 kB
-
https://huggingface.co/PYTHAI/Kimi-K3-fork/resolve/main/FORK.json
- Command line
-
hf download hf://PYTHAI/Kimi-K3-fork/FORK.json
-
curl -L -o FORK.json https://huggingface.co/PYTHAI/Kimi-K3-fork/resolve/main/FORK.json
3.11 kB
| { | |
| "kind": "licence-locked pointer fork", | |
| "source_repo": "moonshotai/Kimi-K3", | |
| "source_url": "https://huggingface.co/moonshotai/Kimi-K3", | |
| "source_commit": "f831ab66814297da540d832a5235f8e904f29d06", | |
| "forked_at_utc": "2026-09-13T23:55:35Z", | |
| "licence_tag": "other", | |
| "licence_file": { | |
| "path": "LICENSE", | |
| "bytes": 3065, | |
| "sha256": "20c797ce19af0c17de52c6afb144644768a591c521655f5ebf5712c9850f2887" | |
| }, | |
| "weights_copied": false, | |
| "weights_not_copied": { | |
| "files": 97, | |
| "bytes": 1560995855544 | |
| }, | |
| "parameters": 2779931837184, | |
| "load_weights_with": "from_pretrained('moonshotai/Kimi-K3', revision='f831ab66814297da540d832a5235f8e904f29d06')", | |
| "files": [ | |
| { | |
| "path": "LICENSE", | |
| "bytes": 3065, | |
| "sha256": "20c797ce19af0c17de52c6afb144644768a591c521655f5ebf5712c9850f2887" | |
| }, | |
| { | |
| "path": "README.md", | |
| "bytes": 45261, | |
| "sha256": "57de265b5842dfa465c6e73b368b0e15a89b8793b5450528dad577da202cc6fe" | |
| }, | |
| { | |
| "path": "assets/kimi-logo.png", | |
| "bytes": 87988, | |
| "sha256": "4a80f64242bf907940765adc7bcf340c28dd83334b07c5503792a26495d1933b" | |
| }, | |
| { | |
| "path": "config.json", | |
| "bytes": 7006, | |
| "sha256": "9710e121a58d03ac92c8d6da287a19541994319afbbe6d6202af001ffd379213" | |
| }, | |
| { | |
| "path": "configuration_kimi_k3.py", | |
| "bytes": 11343, | |
| "sha256": "735eb9ebe593e17d231e08e1df7f7be9b5ee0e079f511aa201f9572077b416ae" | |
| }, | |
| { | |
| "path": "encoding_k3.py", | |
| "bytes": 26304, | |
| "sha256": "49ff03305fdc4be26867972788d36150b67f8a9e852e62bb7959d87482223676" | |
| }, | |
| { | |
| "path": "generation_config.json", | |
| "bytes": 53, | |
| "sha256": "c6648c25e9705af7fba8847e243840d21b5cc63ddeb6297f750a7ddbb6a02836" | |
| }, | |
| { | |
| "path": "kimi_k3_processor.py", | |
| "bytes": 7660, | |
| "sha256": "ec9f7e86d2ab0eee07a8e7e7c037046e77ac3c25a710ad1298ec13be3b585b54" | |
| }, | |
| { | |
| "path": "kimi_k3_vision_processing.py", | |
| "bytes": 6686, | |
| "sha256": "d122b30bfd3a51a6f05d4bfcfda1e657827322b1353f7caefeebc2835d7736b5" | |
| }, | |
| { | |
| "path": "media_utils.py", | |
| "bytes": 13844, | |
| "sha256": "78403540328f9847d6b7ebc5c44eb2e6a752863de0afb7d0710728bb161dc60d" | |
| }, | |
| { | |
| "path": "modeling_kimi_k3.py", | |
| "bytes": 53444, | |
| "sha256": "b9171c96726eda55234c92ac8dfae7e24c512fda68968ae8f2c3782b42665ea2" | |
| }, | |
| { | |
| "path": "modeling_kimi_linear.py", | |
| "bytes": 51506, | |
| "sha256": "9e3564c70ac21854ce5a090cc946c5dc76b70d1050ef50840449181a20fff44a" | |
| }, | |
| { | |
| "path": "preprocessor_config.json", | |
| "bytes": 1011, | |
| "sha256": "4be333605990c53a816e586dee9d5dd545afb7a59947c17f8f7ef26b4782668e" | |
| }, | |
| { | |
| "path": "tiktoken.model", | |
| "bytes": 2795286, | |
| "sha256": "b6c497a7469b33ced9c38afb1ad6e47f03f5e5dc05f15930799210ec050c5103" | |
| }, | |
| { | |
| "path": "tokenization_kimi.py", | |
| "bytes": 16145, | |
| "sha256": "f28ea66e2d862a2a5814970b2ce40c2f7d8296ff09aed90a7e7def689b906944" | |
| }, | |
| { | |
| "path": "tokenizer_config.json", | |
| "bytes": 3478, | |
| "sha256": "5d0803c94db9cd78763499e0956c95fd5a225c14a727e5a6cf5db3f96f010a6e" | |
| } | |
| ], | |
| "why": "preserve the licence terms of this exact commit and pin the revision; the weights stay at the source and are addressed by commit sha", | |
| "forked_by": "mindX (PYTHAI) via the node token" | |
| } |