LFM-7B: benchmark results
Provider: Liquid AI. Released 2025-01-20. Access: API.
Unified ELO 1432 ± 31, rank #1646 of 2656 rated models, from 25 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AI Chess Leaderboard (Continuation) | 795 | Elo | 70.7 |
| AI Chess Leaderboard (Reasoning) | 550 | Elo | 21.1 |
| AI for Education Pedagogy - Primary | 68.08 | Accuracy (%) | 19.5 |
| AI for Education Pedagogy - Maths | 57.94 | Accuracy (%) | 18.2 |
| AI for Education Pedagogy | 61.85 | Accuracy (%) | 18 |
| AI for Education Pedagogy - Social studies | 59.09 | Accuracy (%) | 17.2 |
| AI for Education Pedagogy - Secondary | 61.01 | Accuracy (%) | 16.8 |
| AI for Education Pedagogy - Technology | 67.92 | Accuracy (%) | 16.8 |
| U-MATH - Algebra | 56.67 | Accuracy (%) | 15.2 |
| U-MATH - Precalculus | 56.25 | Accuracy (%) | 15.2 |
| AI for Education Pedagogy - Science | 56.28 | Accuracy (%) | 12.3 |
| BenchTable - Utility | 23.1 | Weighted Score (%) | 11.5 |
Interactive version: theaggregate.ai/model?slug=lfm-7b · How It Works · Data refreshed daily, snapshot 2026-09-19.