Tencent Hunyuan’s Hy3 ranks #25 in Agent Arena and #2 among open models in Frontend Code Arena
arena · x · 2026-07-22
Tencent Hunyuan’s Hy3 lands at #25 overall in Agent Arena
Tencent Hunyuan’s Hy3 is now ranked #25 overall in Agent Arena for open-weight models, with a #5 spot among open-weight models. It also reaches #2 among open models in the Frontend Code Arena, where it places #16 overall.
The post highlights a split profile:
- Strengths: tool use, especially recovery from CLI/bash errors (+2.6%, #25)
- Weakness: steerability, where it struggles to course-correct when users push back (-7.1%, #30)
Agent Arena evaluates models on millions of real-world, long-horizon tasks using web search, filesystem, and terminal tools for workflows like coding, slide decks, web research, app building, and document analysis.
More from coding & agent
- Agent verified 'success' after double-voting: a lesson in harness vs model bugs — TejasKumar_ · 2026-10-26
- diginomica: agentic automation hype is overdone, and 'The Hackening' of enterprise security has begun — jonerp · 2026-10-26
- 'Vibe verification': ~60% of AI-written formal specs prove the wrong property — basedjensen · 2026-10-26
- Devin launches Security suite powered by Claude Mythos with $100M+ credits — DevinAI · 2026-10-26
- HackBench injects reward hacks into agent trajectories to stress-test monitors — DanielKhashabi · 2026-10-26
- Hugging Face ships hf CLI updates: deduped cache saves 19GB, faster buckets, retryable GPU jobs — yb2698 · 2026-10-26