GameFeel Bench v1.4: Qwen-4-Flash-Early leads at 49.9 after score fix
Sea_Emergency_8458 · reddit · 2026-11-09
A Reddit user shared corrected data from the public GameFeel Bench v1.4 (kodasdream.github.io). The widely circulated chart was pre-fix: the benchmark author had miscalculated Qwen-4-27B-Early's audio score, understating its overall from 47 to 44.6. Current leaders are Qwen-4-Flash-Early at 49.9 and Opus-5.5 XHigh at 47.2. It's a niche single-author game-dev benchmark testing early unreleased Qwen 4 checkpoints, so treat results as informal.
More from Models
- Thinky Inkling 2 Small: New Open US Model Trails DeepSeek Flash While Costing More — intellectronica · 2026-11-09
- Gemini 4 Carbon Ranks #1 on Artificial Analysis but No One Can Actually Use It — Yuchenj_UW · 2026-11-09
- Users report inconsistent AI bot allowance usage and weekend performance dips — mattyp · 2026-11-09
- Uncensored 9B distill 'DeepSeek-V4-Pro-Qwen3.5-9B' claims to run locally on a 5GB laptop — udmrzn · 2026-11-09
- Garry Tan's open-source eval: 7 coding harnesses tie on capability but differ 2x on cost — garrytan · 2026-11-09
- Researchers question AI capability acceleration: shift to RL may be the confounder — DavidSKrueger · 2026-11-09