Kimi-K25-eagle3

Eagle3 for Kimi-K2.6 and Kimi-K2.7-Code

Kimi-K2.6 https://huggingface.co/AQ-MedAI/Kimi-K26-eagle3

Kimi-K2.7-Code https://huggingface.co/AQ-MedAI/Kimi-K2.7-Code-eagle3

Eagle3 v2 Optimized Release for Kimi-K2.5 (Build 20260329)

Thanks to the SpecForge framework for their foundational contributions. Stay tuned for further updates.

Model Overview

Kimi-K25-eagle3 is an advanced and highly specialized draft model meticulously engineered to significantly accelerate the inference process of the Kimi-K25 ecosystem, leveraging the powerful EAGLE3 framework.

Architected upon the robust Llama architecture, this model functions as an exceptionally efficient drafter. It has undergone rigorous training on 2 million high-quality samples sourced from the comprehensive EagleChat, open-perfectblend dataset and some multimodal data. This extensive training ensures precise and strict alignment with the teacher model's distribution, thereby guaranteeing high fidelity and performance.

Performance & Acceleration

The core value of this EAGLE3 model is its ability to predict multiple future tokens that are subsequently verified by the base model. High acceptance lengths indicate significant latency reduction. Continuous future iterations.

Speculative Decoding Configuration:

• --speculative-num-steps 3: Configures the number of speculative decoding steps.
• --speculative-eagle-topk 1: Sets the top-k value for the Eagle draft model during speculative decoding.
• --speculative-num-draft-tokens 4: Specifies the number of draft tokens generated in each speculative step.

Average Token Acceptance Lengths:

These metrics demonstrate robust acceleration performance across diverse and complex domains.

Quick Start

Requirements

• NVIDIA GPU
• CUDA 12.0+
• PyTorch 2.0+

Installation

pip install sglang==0.5.9

and include PR https://github.com/sgl-project/sglang/pull/19689

Inference with SGLang

python3 -m sglang.launch_server  \
    --model-path /models/Kimi-K25 \
    --host 0.0.0.0 --port 30012  \
    --trust-remote-code  \
    --mem-fraction-static 0.9 \
    --tp-size 8  \
    --speculative-algorithm EAGLE3  \
    --speculative-draft-model-path AQ-MedAI/Kimi-K25-eagle3 \
    --speculative-num-steps 3  \
    --speculative-eagle-topk 1   \
    --speculative-num-draft-tokens 4

Citation

If you use this model in your research or application, please cite the following:

@misc{kimik25eagle3,
  title={Kimi-K25-eagle3: Accelerating Instruction Following with EAGLE},
  author={Ant AQ Team},
  year={2026},
}