Tencent Open-Sources Hunyuan Hy3 for Agentic Workloads
Multiple posts say Tencent has released and open-sourced the formal version of Hunyuan Hy3, a 295B-parameter MoE model with 21B active parameters and 256K context. It stands out because it targets agentic workflows, tool use, coding, and long-context reasoning while keeping active compute relatively low, and it quickly landed in ecosystem products and inference stacks.
Key details
According to testingcatalog and other posters, Hy3 is the formal release that follows an April-end Hy3 Preview. Posts describe it as a 295B MoE model with 21B active parameters; testingcatalog also mentions a 3.8B MTP layer. Multiple posts add that it is released under Apache 2.0 and can be used commercially.
AdinaYakup says the model supports switchable reasoning modes, including no-thinking, low-intensity, and high-intensity settings, and that FP8 variants are available. aigclink further relays claimed reliability improvements over the previous generation, including hallucination rate dropping from 12.5% to 5.4%, commonsense error rate from 25.4% to 12.7%, and multi-turn failure rate from 17.4% to 7.9%. Because these figures are relayed by third-party posts in this cluster, they should be treated as reported claims here.
Several posts also highlight Tencent's own comparative positioning. iamfakhrealam says Tencent claims Hy3 can compete with open flagship models 2 to 5 times its size, and that in an internal blind test involving 270 experts on real work tasks it scored above GLM-5.1, with the largest gains in tool use, multi-step tasks, and web search.
Availability and ecosystem
vLLM says it natively supported Hy3 on day one, including tool-call parsing, reasoning parsing, and MTP speculative decoding, with validation on both NVIDIA and AMD hardware. Nous Research says Hy3 is available for free on Nous Portal for two weeks. Other posters add that the model is also accessible on OpenRouter, with iamfakhrealam reporting pricing at RMB 1 per million input tokens, RMB 4 per million output tokens, and RMB 0.25 for cache hits.
Reactions and early impressions
Discussion around Hy3 focused not only on benchmark scores but also on efficiency. kimmonismus notes that the model reportedly reaches its efficiency using standard GQA rather than sparse attention or MLA, arguing that this suggests further architectural optimization room still exists. XianbaoQIAN says that in personal testing the model delivered strong opencode performance and could be especially competitive for deployments on 8x H200 servers.
rohanpaulai separately cites an atomic.chat local physics-simulation test in which Hy3 achieved physics quality close to Gemini 3.5 at roughly 35x lower cost. Sam Witteveen frames Hy3 as a serious new competitor to GLM 5.2, especially because it appears strong on many non-coding tasks while using about half the active parameters.
2026-07-06 ~ 2026-07-08 · 26 related posts
Primary sources
- Tencent Hunyuan Hy3 Officially Released with Day-One Native vLLM Support — vllm_project ·
- Tencent Hunyuan Hy3 295B MoE Free on Nous Portal for Two Weeks — NousResearch ·
- Tencent Hunyuan Hy3 Open-Sourced: 295B Total Params, 21B Activated, Apache 2.0 — eliebakouch · 2026-07-06
- Hy3 Beats GLM-5.1 in Blind Tests, Excels in Frontend and CI/CD — eliebakouch · 2026-07-06
- Tencent Hy3 Tested: Half the Active Params of GLM 5.2, Runs Efficiently on 8x H200 — Xianbao_QIAN · 2026-07-06
- ModelScope Releases Hy3: 295B Total Param MoE Model with 256K Context — bdsqlsz · 2026-07-06
- Tencent Hunyuan v3 (Hy3) MoE Hits HF Trending — tencent · 2026-07-06
- Tencent Open-Sources Hunyuan Hy3: 295B MoE, Hallucination Rate Drops to 5.4% — aigclink · 2026-07-06
- Tencent Open-Sources Hunyuan HY3: 295B MoE with Switchable Reasoning — AdinaYakup · 2026-07-06
- Tencent Hunyuan Hy3 Released with Focus on Real-World Workflows — airesearch12 · 2026-07-06
- Tencent Open-Sources Hy3: 295B MoE Model Rivals Larger Flagships — yshan2u · 2026-07-06
- Tencent Open-Sources Hunyuan Hy3 for Coding and Agents — 智东西 · 2026-07-06
- Tencent Launches Hunyuan 3: 295B MoE Beats GLM-5.1 in Blind Test — kimmonismus · 2026-07-06
- Tencent Releases Hy3 Open-Source LLM Official Version — testingcatalog · 2026-07-06
- Tencent Hunyuan Hy3 Launched: 295B MoE Model for Coding and Agents — AGI Hunt · 2026-07-06
- [source] Tencent Hunyuan Hy3 Officially Released with Day-One Native vLLM Support — vllm_project · 2026-07-06
- Tencent Hunyuan Hy3 Matches Gemini 3.5 in Local Physics Simulation — rohanpaul_ai · 2026-07-07
- [source] Tencent Hunyuan Hy3 295B MoE Free on Nous Portal for Two Weeks — NousResearch · 2026-07-07
- Nous Launches Portal Subscription: One-Stop Hermes Agent Building — NousResearch · 2026-07-07
- Tencent Releases Hy3 Model — bdsqlsz · 2026-07-07
- Tencent Hunyuan Hy3 Officially Released with Upgraded Yuanbao Agent Capabilities — 量子位 · 2026-07-07
- Tencent Hy3 Explained: A New Rival to GLM 5.2 — Sam Witteveen · 2026-07-07
- Tencent Hunyuan Hy3 Nears Claude Opus 4.8 Level — DeryaTR_ · 2026-07-07
- Tencent Releases Open-Source MoE Model HY3 — iamfakhrealam · 2026-07-07
- Tencent's New Model Claims to Surpass GLM-5.1 — iamfakhrealam · 2026-07-07
- Aggressive Pricing for Tencent's New Model: 1 RMB/Million Input Tokens — iamfakhrealam · 2026-07-07
- Tencent Hunyuan HY3 Released: 295B MoE for Coding and Automation — iamfakhrealam · 2026-07-07
1 near-duplicate retellings: rohanpaul_ai