HY4 Shows Huge Capability Jump Over Preview, Wins Praise for Honest Evals
Xianbao_QIAN · x · 2026-11-04
Xianbao QIAN highlights the HY4 release, praising the HY series for being unusually honest on evals, with work by Shunyu Yao. A 12-benchmark chart shows Hy4 leading on coding and agent evals like ProgramBench, FrontierCode 2.0, Terminal Bench 4.0 and OSWorld 2.1 against GLM 5.3, DeepSeek-V4.1-Flash, Kimi K3.1 Pro, Claude Opus 5.5 and GPT 6.1 Sol, marking a big jump from Hy4-preview.
More from Models
- Hobbyist rewrites Qwen 0.6B weights with 39.8M tokens, no gradients, in 3h12m — Altruistic_Heat_9531 · 2026-11-04
- Same leak, reshared: ChatGPT's five family plan prices spotted again — heyshrutimishra · 2026-11-04
- KOL visits StepFun's Shanghai office, hints a powerful new model is coming — teortaxesTex · 2026-11-04
- Fable 5.5 delayed as Anthropic reportedly bets OpenAI's Astra 6.2 won't beat it — haider1 · 2026-11-04
- ByteDance open-sources Axion 1.0-35B agent foundation model with full reproducible recipe — pmttyji · 2026-11-04
- Community asks Moonshot when Kimi K3.1 model weights will drop — bgmshana · 2026-11-04