Tencent Releases Hy4 Preview: 770B Params, 1M Context, Reasoning Mode
Simon Willison · rss · 2026-08-30
Tencent released Hy4 Preview, a new open-weight text LLM. It features 770B total parameters (49B active), a 1M token context window, and weighs 1.56TB on Hugging Face—a significant scale-up from the July Hy3 (295B/256k). The config reveals two reasoning effort levels: 'high' (default) and 'nothink'. Tests show the hidden reasoning trace optimizes for token efficiency, using truncated English when debating details like adding a helmet to a pelican.
More from Models
- OpenAI stealth-testing mysterious internal model 'Aeon' in Codex desktop app and Dots — cedric_chee · 2026-10-16
- Argon edges out astra as daily driver, but Claude still wins heavy coding work — iruletheworldmo · 2026-10-16
- NVIDIA ships FP4-quantized Qwen3.8-Flash for local use, trending on Hugging Face — nvidia · 2026-10-16
- RoboVista benchmark: Gemini-4 Argon hits 81.9% on robotics understanding — berkeley_ai · 2026-10-16
- Dev reports decision model jev beats gpt-6-luna on quality in real training pipeline — genmon · 2026-10-16
- Google won't publish Argon 4 limits in AI Pro—users compare experiences — Consistent-Text-2662 · 2026-10-16