RedHatAI / Kimi-K2.6-NVFP4

huggingface.co
Total runs: 17.3K
24-hour runs: 0
7-day runs: 9.3K
30-day runs: 14.3K
Model's Last Updated: May 01 2026

Introduction of Kimi-K2.6-NVFP4

Model Details of Kimi-K2.6-NVFP4

NVFP4-Quantized RedHatAI/Kimi-K2.6-NVFP4

This is a preliminary version (and subject to change) of NVFP4-quantized moonshotai/Kimi-K2.6 model, for performant inference on NVIDIA Blackwell GPUs. The model has both weights and activations quantized in NVFP4 format with vllm-project/llm-compressor .

It is compatible and tested against vllm v0.20.0. Deploy it via vllm serve using the recipes at https://recipes.vllm.ai/moonshotai/Kimi-K2.6 .

Creation Script:

Kimi K2.6 support will land in https://github.com/vllm-project/llm-compressor/pull/2662 . The script to create will be posted as a link shortly

Preliminary Evaluations

  1. GSM8K Platinum:
lm_eval --model local-chat-completions \
  --tasks gsm8k_platinum_cot_llama \
  --model_args "model=RedHatAI/Kimi-K2.6-NVFP4,max_length=262144,base_url=http://0.0.0.0:8000/v1/chat/completions,num_concurrent=128,max_retries=3,tokenized_requests=False,tokenizer_backend=None,timeout=1200" \
  --num_fewshot 0 \
  --apply_chat_template \
  --gen_kwargs "do_sample=True,temperature=1.0,top_p=0.95,top_k=20,min_p=0.0,max_gen_toks=64000,presence_penalty=1.5,repetition_penalty=1.0,seed=5678"

Recovery:

moonshotai/Kimi-K2.6
(original in W4A16)
RedHatAI/Kimi-K2.6-NVFP4
(this model)
Accuracy
94.29 93.96
Recovery - 99.6%

Note : More rigorous evaluations are currently in progress and will be available soon.

Runs of RedHatAI Kimi-K2.6-NVFP4 on huggingface.co

17.3K
Total runs
0
24-hour runs
0
3-day runs
9.3K
7-day runs
14.3K
30-day runs

More Information About Kimi-K2.6-NVFP4 huggingface.co Model

Kimi-K2.6-NVFP4 huggingface.co

Kimi-K2.6-NVFP4 huggingface.co is an AI model on huggingface.co that provides Kimi-K2.6-NVFP4's model effect (), which can be used instantly with this RedHatAI Kimi-K2.6-NVFP4 model. huggingface.co supports a free trial of the Kimi-K2.6-NVFP4 model, and also provides paid use of the Kimi-K2.6-NVFP4. Support call Kimi-K2.6-NVFP4 model through api, including Node.js, Python, http.

Kimi-K2.6-NVFP4 huggingface.co Url

https://huggingface.co/RedHatAI/Kimi-K2.6-NVFP4

RedHatAI Kimi-K2.6-NVFP4 online free

Kimi-K2.6-NVFP4 huggingface.co is an online trial and call api platform, which integrates Kimi-K2.6-NVFP4's modeling effects, including api services, and provides a free online trial of Kimi-K2.6-NVFP4, you can try Kimi-K2.6-NVFP4 online for free by clicking the link below.

RedHatAI Kimi-K2.6-NVFP4 online free url in huggingface.co:

https://huggingface.co/RedHatAI/Kimi-K2.6-NVFP4

Kimi-K2.6-NVFP4 install

Kimi-K2.6-NVFP4 is an open source model from GitHub that offers a free installation service, and any user can find Kimi-K2.6-NVFP4 on GitHub to install. At the same time, huggingface.co provides the effect of Kimi-K2.6-NVFP4 install, users can directly use Kimi-K2.6-NVFP4 installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

Kimi-K2.6-NVFP4 install url in huggingface.co:

https://huggingface.co/RedHatAI/Kimi-K2.6-NVFP4

Url of Kimi-K2.6-NVFP4

Kimi-K2.6-NVFP4 huggingface.co Url

Provider of Kimi-K2.6-NVFP4 huggingface.co

RedHatAI
ORGANIZATIONS

Other API from RedHatAI

huggingface.co

Total runs: 26.2K
Run Growth: -15.7K
Growth Rate: -59.90%
Updated:May 18 2026
huggingface.co

Total runs: 20.9K
Run Growth: -11.8K
Growth Rate: -56.48%
Updated:April 29 2026