François Chollet explains ARC-3 evaluation on Kaggle
fchollet · x · 2026-08-31
François Chollet clarifies that the ARC 3 competition on Kaggle runs on a private test set. A separate evaluation process is used for frontier model APIs, run internally when requested by partners.
Related event: Chollet Responds to ARC-AGI Benchmark Controversy(3 posts)→
More from Models
- Tiel-Coder-35B-A3B Trends on HF with Speculative Decoding & MTP — peculiar-ragdoll · 2026-08-31
- Google Releases Gemini Omni 1.1 Flash, Updating Its Fast Multimodal Model for Developers — thione · 2026-08-31
- DeepSeek launches low-cost vision model; Anthropic previews hardware control protocol for agents — thione · 2026-08-31
- Qwen and GLM release new MoE models focusing on low cost and high performance — thione · 2026-08-31
- Professor says Claude grammar fixes got his writing flagged 100% AI by Pangram — ipeirotis · 2026-08-31
- Anthropic: Claude-powered automated alignment researchers beat veteran humans' ideas — burny_tech · 2026-08-31