Arcee-Blitz — benchmark results

Arcee AI's 24B Mistral Small 3-based workhorse model, distilled from DeepSeek-V3. Provider: Arcee AI. Released 2025-02-07. Access: Open.

Unified ELO 1564 ± 30, rank #581 of 1776 rated models, from 10 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard - MMLU-Pro57.26Score99.9
Open LLM Leaderboard - MuSR23.82Score98.6
Open LLM Leaderboard - GPQA18.01Score96.2
Open LLM Leaderboard - BBH50.73Score93.4
Open LLM Leaderboard - MATH Level 534.82Score85.2
Open LLM Leaderboard - IFEval55.43Score67
UGI - Willingness (W/10)6.8W/10 Score66.6
UGI - Natural Intelligence23.14NatInt Score55.5
UGI Leaderboard32.57UGI Score41.6
UGI - Writing23.26Writing Score18.9

Interactive version: theaggregate.ai/model?slug=arcee-blitz · How the rankings work · Data refreshed daily, snapshot 2026-07-22.