Child-safety benchmark KORA spurs real gains: Mistral jumps from last to 10th of 42
GregKamradt · x · 2026-10-15
The KORA child-safety AI benchmark, built by Mathilde's team, has achieved something rare: direct model improvement. Mistral massively improved its child-safety score in a recent release — jumping from last place to 10th out of 42 models — and cited KORA on its official model card. ARC Prize's Greg Kamradt notes benchmarks rarely drive improvements this directly.
More from Models
- Cursor + Gemini 4 Argon builds a multiplayer 3D capture-the-flag game in one go — FinanceYF5 · 2026-10-15
- Claude Opus 5.5 levels up codegen but its docs and PR prose are 'unusable' — indragie · 2026-10-15
- SFT learns arithmetic mixture, OPD learns geometric mean: why distillation forgets differently — QuanquanGu · 2026-10-15
- GPT-6 partially solves NIST assembly board zero-shot for just $3.98 — DJiafei · 2026-10-15
- vLLM ships Day-0 support for Step 5 Preview, cuts sparse prefill latency 2.2% on B300 — vllm_project · 2026-10-15
- vLLM details Step 5 Preview's region-based sparse attention: score summaries, then attend — vllm_project · 2026-10-15