# Tencent-Hunyuan/HunyuanImage-3.0

HunyuanImage-3.0: A Powerful Native Multimodal Model for Image Generation

Repository: https://github.com/Tencent-Hunyuan/HunyuanImage-3.0
Canonical: https://ross.abutalabs.com/products/hunyuanimage-30
Homepage: https://hunyuan.tencent.com/image
Language: Python
License: NOASSERTION
License Family: other
Topics: image-generation, native-multimodal-model
Last push: 2026-06-23T13:12:44+00:00

## Health v2 (maintenance only)
Score: 57/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 89, release rhythm 35, longevity 24
- inputs: {"age_days": 340, "days_push": 71, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3253, forks 186 (observed 2026-08-28T04:07:52.167616+00:00)

## What it is
HunyuanImage-3.0 is Tencent's open-source native multimodal model for text-to-image and image-to-image generation, with inference code and model weights on HuggingFace. It includes an Instruct variant with reasoning-based prompt enhancement and a distilled checkpoint for faster 8-step sampling, plus vLLM-accelerated inference.

## Use cases
- generate images from text prompts
- edit images with instruction-based image-to-image generation
- run a powerful open-source text-to-image model locally on GPUs
- speed up image generation inference with vLLM
- use a distilled model for fast few-step image sampling
- enhance prompts automatically with a reasoning-capable image model

## When to choose
- you need state-of-the-art open-weight text-to-image generation
- you want instruction-based image editing in one multimodal model
- you have GPU resources and want local, self-hosted image generation
- you need fast inference via distilled checkpoints or vLLM

## When to avoid
- you lack a high-memory GPU, as the model is very large
- you need a permissively licensed model for commercial embedding without checking the custom license
- you only need lightweight or CPU-only image generation

## Facets
- artifact type: library
- maturity: active
- function: image-processing, machine-learning, deep-learning, llm-inference
- domain: artificial-intelligence, image-processing, deep-learning, large-language-models
- platform: python
- tags: image-generation, text-to-image, multimodal-model, diffusion, vllm, huggingface, gpu, linux

## Member repositories
- Tencent-Hunyuan/HunyuanImage-3.0 (main) score 57

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:52.167616+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:24:24.096094+00:00, confidence not recorded.
  - readme: https://github.com/Tencent-Hunyuan/HunyuanImage-3.0 (fetched 2026-08-28T04:07:52.167616+00:00, sha b1067cbbef25)
  - homepage: https://hunyuan.tencent.com/image (fetched 2026-08-29T09:36:56.744121+00:00, sha e9ec7b1723ad)
- Data as of 2026-08-30T08:39:29.467469+00:00.