AITraining2U

Programs

Resources

Case Studies

Quick Links

Enquire Now
Open-Source Models

Tencent Hunyuan Hy3: efficient MoE, strong multimodal

Tencent's Hunyuan Hy3 does a lot with a little: a 295-billion-parameter mixture-of-experts that fires only 21B parameters per token, Apache-2.0 licensed, and unusually good at image, 3D and video generation.

By AITraining2U Editorial Team 2026-07-18 8 min read
Tencent Hunyuan multimodal model generating image and 3D

Tencent's Hunyuan family rounds out the 2026 Chinese open-model field, and its angle is efficiency plus multimodality. Where others chase raw scale, Hunyuan Hy3 is engineered to match much bigger models on agentic and coding tasks while activating only a small slice of its parameters — and Tencent brings serious pedigree in image, 3D and video generation.

Model: Hunyuan Hy3 / HunYuan 3.0 (July 2026) • Params: 295B total MoE, 21B active • Context: 256K tokens • Licence: Apache-2.0 • Extra strengths: Image, 3D and video generation • Access: Hugging Face, Tencent Cloud

A sparse model that punches up

Hy3 is a 295-billion-parameter mixture-of-experts that routes each token through just 21 billion active parameters. That 14-to-1 sparsity is the point: Tencent positions Hy3 as matching the agentic and coding performance of flagship models 2–5× its size, at far lower cost per token. If you want the intuition for how a big model can be cheap to run, our parameters & scaling laws guide explains MoE sparsity.

Apache-2.0 and a 256K context

Hy3 is released under Apache-2.0 — permissive, commercial-friendly, self-hostable — with a 256K-token context window. That is smaller than the 1M-token contexts of Kimi K3, DeepSeek V4 and GLM-5.2, but ample for most agent and document tasks, and the smaller active footprint makes it comparatively affordable to serve.

The multimodal edge

Tencent's differentiator is generative multimodality. Beyond text, the Hunyuan ecosystem includes strong image, 3D-asset and video generation models (the lineage behind HunyuanVideo and Hunyuan3D). For teams whose work is visual — marketing content, game assets, product imagery, short video — Hunyuan is often the most directly useful open family, not just a text model with pictures bolted on.

Where Hunyuan fits for a Malaysian team

Two sweet spots. First, cost-efficient agents: if you want strong coding/agent behaviour without the GPU bill of a 700B-plus model, Hy3's sparse design is attractive and Apache-2.0 keeps it self-hostable for PDPA control. Second, creative and multimodal production: image, 3D and video generation for content, e-commerce and media. Build either on the deployment skills in our AI Engineering course, and orchestrate multi-step pipelines with n8n.

Limitations

The 256K context, while generous, trails the 1M-token leaders for very long single-request tasks. On the hardest pure-reasoning benchmarks the largest models may still edge ahead. And as with the multimodal generators, check licensing and usage terms per component — multimodal releases sometimes carry conditions the text model does not. Validate on your workload and pin the version.

Hunyuan Hy3 benchmarks (2026)

BenchmarkScoreWhat it measures
SWE-bench Verified74.4%Real GitHub issue fixes — exceeds most closed models
GPQA Diamond90.4Graduate-level science reasoning
USAMO 202672.0Olympiad-level mathematics
Expert blind eval2.67 / 4270 experts on real workplace tasks (vs GLM-5.1's 2.51)
Tencent-reported (July 2026) for the Apache-2.0 Hy3 release.

Official resources & downloads

Go straight to the source — official sites, model cards and weight downloads:

Sources & Further Reading

Specs and benchmarks reflect vendor announcements and independent leaderboards current as of July 2026; fast-moving — verify the latest figures before production decisions.

Frequently Asked Questions

Hunyuan Hy3 (HunYuan 3.0, July 2026) is Tencent's open-weight flagship — a 295-billion-parameter mixture-of-experts model that activates just 21B parameters per token, released under Apache-2.0 with a 256K-token context. Tencent positions it as matching models 2–5× its size on agentic and coding tasks.

Yes — the text model is released under Apache-2.0, a permissive licence allowing commercial use and self-hosting. Tencent's separate multimodal generators (image, 3D, video) may carry their own terms, so check each component's licence.

Its mixture-of-experts design routes each token through only 21B of its 295B parameters — a roughly 14-to-1 sparsity — so it runs far cheaper than a dense model of similar capability, while still matching much larger models on many tasks.

Yes — multimodal generation is Tencent's differentiator. The Hunyuan ecosystem includes strong image, 3D-asset and video generation models, making it especially useful for marketing content, game assets, product imagery and short video.

Its 256K-token context is generous and enough for most agent and document work, but smaller than the 1M-token windows of Kimi K3, DeepSeek V4 and GLM-5.2. For very long single-request tasks, those models have more headroom.

Build efficient, multimodal AI systems

Learn to deploy and orchestrate open models — text and multimodal — in production. Our AI Engineering course is HRDC-claimable for eligible Malaysian employers.