Testing Tencent Hunyuan Hy4 preview: 17/105 bugs fixed for just $3.13
PawelHuryn · x · 2026-08-28
The author planted 105 hard bugs across 2 real repos and asked models to find and fix them—labs don't know the answer key, making it closer to real-world use.
Results:
- Fable 5 (max): 29/105 for $104.49
- Opus 5 (max): 27/105 for $51.33
- Qwen3.8-Max (xhigh): 18/105 for $31.19
- Hy4 preview (default): 17/105 for just $3.13
- Gemini 3.7 Flash (high): 16/105 for $6.36
Verdict: not a breakthrough, but Hy4 sits on the price-performance Pareto frontier with open weights—a solid everyday coding partner.
Related event: Tencent Hunyuan Hy4 Fixes 17 Bugs for $3.13 in Real-Repo Blind Test(4 posts)→
More from coding & agent
- YC F26's Praxis launches AI agents to run law firm ops end to end — ycombinator · 2026-10-16
- Codex's year-long rise: 40M+ weekly users and its leader now runs all of OpenAI's core products — danshipper · 2026-10-16
- Developer builds inter-agent messaging for his AI agents — complete with message undo — baditaflorin · 2026-10-16
- How Codex Happened: inside OpenAI's comeback in coding — danshipper · 2026-10-16
- iOS 'dots' workflow: one Siri command turns screenshots into a background shopping and saving agent — nickbaumann_ · 2026-10-16
- Your widget tree is already a list of options: where Jev fits in Flutter — rseroter · 2026-10-16