Hy4-preview README

The project's README from github.com/Tencent-Hunyuan/Hy4-preview. As one markdown file: README.md.

中文 ｜ English



License    HuggingFace    ModelScope    cnb.cool    GitCode

🖥️ Official Website  |   💬 GitHub

Table of Contents

Model Introduction

Hy4 preview is a new-generation Mixture-of-Experts (MoE) flagship model developed by the Tencent Hy Team. The model comprises 770B total parameters, of which 49B are activated per token. The backbone consists of 78 layers, where the first layer uses a standard dense FFN and the remaining 77 layers replace it with MoE, each containing 256 routed experts and 1 shared expert; every token activates the top-8 routed experts along with the shared expert. In addition to the backbone, 1 native MTP layer (10B total parameters, 0.7B activated) is built in for speculative decoding.

On the architecture side, inspired by DeepSeek and GLM, the attention module employs Gated DeepSeek Sparse Attention (Gated DSA) with IndexCache for cross-layer sparse index reuse. The residual pathway uses iHC (identity Hyper-Connections) to expand inter-layer information flow.

Model Specifications

The table below lists backbone parameters only, excluding the MTP layer.

Property Value
Architecture Mixture-of-Experts (MoE)
Total Parameters 770B
Activated Parameters 49B
Layers 78
Hidden Size 6144
Attention Type Gated DSA
Attention Heads 64
Query Compression Dimension 2048
Key-Value Compression Dimension 512
Indexer Heads / Head Dimension 32 / 128
Indexer top-k 2048
Residual Streams 4
Routed Experts 256
Shared Experts 1
Activated Routed Experts per Token 8
MoE Intermediate Size 2048
FFN Intermediate Size 18432
Context Length 1M
Vocabulary Size 120832

A New Flagship Generation

We scaled Hy4 preview on three fronts: model size, context length, and training data. Stronger pre-training and a substantially larger post-training run compound into another step change in capability — the largest generation-over-generation gain we've measured, and enough to put Hy4 preview at the open-source frontier.

Built for Productivity

We partnered with top experts inside Tencent — such as software engineers, game developers, finance analysts, and security experts — and built training data around the work they ship. The result is a model that gets meaningfully further on the tasks these teams run every day:

Software engineering: Better at understanding, planning, debugging, and verifying long-horizon development tasks, with further gains in the visual taste and interaction quality of front-end work.

Office and analysis: Takes messy context spread across many files and converts it into shareable artifacts — documents, spreadsheets, and presentations — handling data analysis, equations, and financial models with greater precision.

Game development: Turns a single prompt into a playable prototype and works fluently with game engines, so developers can keep refining complex projects over multiple turns.

Scientific research: Stronger understanding, reasoning, and problem-solving on hard research questions, with solid progress across AI research, molecular dynamics, condensed matter physics, and pure mathematics.

We also continue to co-design Hy4 preview with Tencent products like CodeBuddy and WorkBuddy, so that gains in the model show up in the work people actually do with it. To check that, we ran a blind side-by-side evaluation: 163 internal experts rated model outputs on 203 engineering tasks. Hy4 preview came out slightly ahead of both GLM 5.3 (2.99 vs. 2.92 average, 46.8% wins / 12.8% ties / 40.4% losses) and Kimi K3 (2.99 vs. 2.94, 51.2% wins / 7.9% ties / 40.9% losses).

Benchmark Appendix

Known Limitations

This is an early version of Hy4. There is real headroom left in both pre-training and post-training, and we are shipping with known issues — among them, spending longer than necessary reasoning through complex tasks, and a tendency to over-verify its own work. We'll keep iterating quickly on these. As with Hy3 preview, we would rather ship early and hear what breaks — that's what made Hy3 substantially better, and it's how we will get Hy4 right. We will also keep collaborating closely with Tencent's products and in-house experts to push the boundaries of model intelligence while making it more abundant and affordable.

News

Model Name Description Hugging Face ModelScope GitCode CNB
Hy4 preview Instruct model 🤗 Model Model Model Model
Hy4 preview-FP8 FP8 quantized instruct model 🤗 Model Model Model Model

Quickstart

Deploy Hy4 preview with vLLM or SGLang first, then call the OpenAI-compatible API:

from openai import OpenAI

client = OpenAI(base_url="http://127.0.0.1:8000/v1", api_key="EMPTY")

response = client.chat.completions.create(
    model="hy4-preview",
    messages=[
        {"role": "user", "content": "Hello! Can you briefly introduce yourself?"},
    ],
    temperature=0.9,
    top_p=1.0,
)
print(response.choices[0].message.content)

Recommended parameters: temperature=0.9, top_p=1.0.

Reasoning mode: Defaults to "high" (deep chain-of-thought), which suits complex tasks such as math, coding, and reasoning. For direct responses, pass extra_body={"chat_template_kwargs": {"reasoning_effort": "no_think"}}.

See the Deployment section below for how to start the API server.

Deployment

For production serving, we recommend using vLLM or SGLang. Please refer to the recipes:

vLLM

Use official prebuilt image vllm/vllm-openai:hy4-preview:

docker run --gpus all \
  -p 8000:8000 \
  --ipc=host \
  -v ~/.cache/huggingface:/root/.cache/huggingface \
  vllm/vllm-openai:hy4-preview tencent/Hy4-preview-FP8 \
    --tensor-parallel-size 8 \
    --speculative-config '{"num_speculative_tokens":3,"method":"mtp"}' \
    --attention-backend FLASHMLA_SPARSE \
    --tool-call-parser hy_v4 \
    --reasoning-parser hy_v4 \
    --enable-auto-tool-choice \
    --port 8000 \
    --served-model-name hy4-preview

SGLang

Use the official prebuilt image lmsysorg/sglang:hy4-preview (multi-arch, x86 and Arm):

docker pull lmsysorg/sglang:hy4-preview

docker run --gpus all --ipc=host -p 8000:8000 lmsysorg/sglang:hy4-preview \
  python3 -m sglang.launch_server \
    --model tencent/Hy4-preview-FP8 \
    --tp-size 8 \
    --reasoning-parser auto \
    --tool-call-parser auto \
    --speculative-algorithm NEXTN \
    --speculative-num-steps 3 \
    --speculative-eagle-topk 1 \
    --speculative-num-draft-tokens 4 \
    --port 8000 \
    --served-model-name hy4-preview

Finetuning

Hy4 preview provides a complete model finetuning pipeline. For detailed documentation, please refer to: Finetuning Guide

Quantization

We provide AngelSlim, a more accessible, comprehensive, and efficient toolkit for large model compression. AngelSlim supports a comprehensive suite of compression tools for large-scale multimodal models, including common quantization algorithms, low-bit quantization, and speculative sampling.

License

Hy4 preview is released under the Apache License 2.0. See LICENSE for details.

Contact Us

If you have any questions or suggestions, feel free to reach out to our R&D and product teams via email:

📧 [email protected]

Hy4 preview is developed by the Tencent Hy Team.

Hy4-preview (Tencent-Hunyuan/Hy4-preview)

long-context
Searching Database…

395 ·GitHub
Hy4-preview
What is this?Hy4 preview is Tencent Hunyuan's newest open-source large language model, a 770B-parameter Mixture-of-Experts model (49B active per token) built for coding, office/data analysis, game development, and scientific research tasks.
Is this a ready-to-use chat app, or just model weights I have to run myself?It is just the model: open-weight instruct and FP8-quantized checkpoints released on Hugging Face, ModelScope, GitCode, and CNB. You have to serve them yourself with vLLM or SGLang and talk to them through an OpenAI-compatible API.
How do I get it running?Pull the official prebuilt vLLM image (vllm/vllm-openai:hy4-preview) or the SGLang image (lmsysorg/sglang:hy4-preview), start the server, then call the model through an OpenAI-compatible API.
What do I need to serve a 770B-parameter model like this?You need a vLLM or SGLang serving stack with enough GPU memory for a 770B total / 49B activated MoE model; the FP8-quantized build and Tencent's AngelSlim toolkit are provided to help reduce the footprint.
What should I expect from this being a 'preview' release?The authors call it an early version with known issues: it sometimes spends longer than necessary reasoning through complex tasks and tends to over-verify its own work, with more training headroom still to come.
How does Hy4 preview compare to other open models?In a blind internal evaluation with 163 experts rating 203 engineering tasks, Hy4 preview slightly edged out GLM 5.3 (46.8% wins vs 40.4% losses) and Kimi K3 (51.2% wins vs 40.9% losses).
Can I fine-tune or quantize the model myself?Yes, Hy4 preview ships with a complete fine-tuning pipeline and Tencent's AngelSlim toolkit for quantization and speculative sampling.
How long a context can it handle?Hy4 preview supports a 1M token context length.
What real-world tasks was it tuned for?Training data was built around work from Tencent software engineers, game developers, finance analysts, and security experts, and the model is co-designed with Tencent products like CodeBuddy and WorkBuddy.
What are similar projects?MiniCPM, DeepSeek-V3, StartLux-Decision, Qwen3.8-Flash-Next, Hy-MT2, Model-Optimizer, ZDTaichu5.0-9B, DeepAnalyze
Which topics is it in?long-context, sglang, foundation-models, inference-engine, mixture-of-experts, reasoning, attention-mechanism, vllm, quantization, fine-tuning, openai-compatible, transformer
Creator
Tencent-Hunyuan
Size
11.6 MB
Stars
395
Updated
CommitAug 28
Age
1mo
Commits · 90 Days