magnum-v4-22B: benchmark results
Anthracite's community magnum v4 fine-tune of Mistral Small Instruct 2409 (22B) aiming to replicate Claude 3 Sonnet/Opus prose quality. Provider: Other. Released 2024-10-20. Access: Open.
Unified ELO 1549 ± 1, rank #443 of 1392 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - GPQA | 10.4 | Score | 79.6 |
| Open LLM Leaderboard - MuSR | 13.43 | Score | 74 |
| Open LLM Leaderboard - BBH | 35.55 | Score | 71.9 |
| Open LLM Leaderboard - MATH Level 5 | 20.02 | Score | 70.9 |
| Open LLM Leaderboard - IFEval | 56.29 | Score | 68.2 |
| Open LLM Leaderboard - MMLU-Pro | 31.44 | Score | 67.1 |
| UGI - Writing | 35.22 | Writing Score | 55.8 |
| UGI - Natural Intelligence | 21.94 | NatInt Score | 50.9 |
| UGI - Willingness (W/10) | 4.8 | W/10 Score | 36.8 |
| UGI Leaderboard | 27.33 | UGI Score | 24.1 |
Interactive version: theaggregate.ai/model?slug=magnum-v4-22b · How It Works · Data refreshed daily, snapshot 2026-09-05.