How we hit #1 on Decision Index Vision: our HF impl capped at 512px
antoine_chaffin · x · 2026-10-09
The author shares how they reached top-1 on the Decision Index Vision: examine the eval data, notice underperformance on tasks needing precise in-image reading, then discover their Hugging Face implementation capped resolution at 512×512 — fix it and profit. @multimodalart then re-ran all models under a comparable 1.6M-pixel cap, quickly improving the leaderboard's fairness.
More from Models
- Datology's Curation Studio claims 6x compute multiplier; Thomson-1 built for $450K — jefrankle · 2026-10-09
- ARC-AGI-3 leader changes: Yi-Chia Chen hits 59.17%, overtaking tufalabs — fchollet · 2026-10-09
- ARC-AGI-2 tops out at 88.06% as tufalabs claims ARC Prize 2026 high score — fchollet · 2026-10-09
- OpenAI's math gains likely from massive Lean-based RL environments; post-training is "rich man's inference" — yacineMTB · 2026-10-09
- Kalshi and Polymarket odds have Claude and Gemini neck-and-neck for best model by EOY on Argon 4 release — matt_slotnick · 2026-10-09
- Codex /fast mode mocked: 16-44 token/s vs Claude Opus 5.5's 80 — 'rename it /wait' — lxfater · 2026-10-09