Mistral's 209M-parameter Timestral claims SOTA on GIFT-Eval time-series forecasting
NielsRogge · x · 2026-10-17
Papers with Code reports Mistral's Timestral-209M achieved SOTA on the GIFT-Eval time-series forecasting benchmark: 0.658 MASE and 0.446 CRPS zero-shot, edging out t0-gamma (0.661 MASE / 0.452 CRPS) and TIME, while also topping fev-bench with a 50.39% skill score. A 209M zero-shot model beating larger specialized forecasters marks a notable small-model milestone.
More from Models
- Saturday model drop beats Opus 5 on agent benchmarks at $0.20 vs $12 per task — sophiamyang · 2026-10-17
- Chartography leaderboard: Claude Sonnet 5.5 tops chart understanding at 0.902 pass@1 — NielsRogge · 2026-10-17
- Pro $200 plan chewed through by daily research automations, user seeks tips — MidnightIndex35 · 2026-10-17
- Delta appears to block Grok bot's IP range, thwarting flight-booking attempt — lucasmeijer · 2026-10-17
- Qwen-Image-Edit shows East Asian facial prior causing identity drift in edits — Gen-Talks · 2026-10-17
- AI turns a nuclear engineer's PWR rod-bow challenge from X into a structured analysis — burhop · 2026-10-17