LiquidAI: lfm-40b
1 benchmarks
Leaderboard position
#1010
Overall rank (of 1121 models)
-125
Elo rating
1
Benchmarks
Top 90.1%
Percentile
In LiquidAI (of 12 models)
Above: Qwen: qwen2.5-vl-7b-instruct #1009
Below: atomicchat: lfm3-1.7b-uncensored #1011
Benchmark highlights
Nearest competitor from another company
The model from another company with the most similar benchmark scores.
appeared as nearest in 1 of 1 benchmarks
1/1
100.0%
Benchmarks
Predicted scores
Estimated from 25 similar models and the global Elo rating. These are projections, not measured results.
| Benchmark | Predicted | Confidence | Based on |
|---|---|---|---|
| MATH-500 | ~69.39 | 50% | — |
| AIME 2025 | ~33.97 | 50% | — |
| AIME 2026 | ~54.55 | 50% | — |
| HMMT Feb 2026 | ~45.99 | 50% | — |
| GPQA Diamond | ~36.53 | 50% | — |
| MMLU-Pro | ~37.19 | 50% | — |
| Humanity's Last Exam (no tools) | ~4 | 50% | — |
| IFEval | ~68.89 | 50% | — |
| IFBench | ~37.12 | 50% | — |
| HumanEval | ~45.27 | 50% | — |
| MBPP | ~39.79 | 50% | — |
| Terminal-Bench v2.1 | ~8.6 | 50% | — |
| MMLU | ~42.43 | 50% | — |
| SWE-bench Verified | ~8.47 | 50% | — |
| SWE-bench Pro | ~5.24 | 50% | — |
| SciCode | ~11.62 | 50% | — |
| AA-LCR | ~23.96 | 50% | — |
| Artificial Analysis Intelligence Index v5.0.1 | ~12 | 50% | — |
| Humanity's Last Exam (with tools) | ~11.9 | 50% | — |
| GSM8K | ~45.98 | 50% | — |
| -(CoderPPL) | ~-6.99 | 50% | — |
| LiveCodeBench v6 | ~33.56 | 50% | — |
| MMMU-Pro | ~34.57 | 50% | — |
| MATH | ~19.65 | 50% | — |
| BBH | ~43.59 | 50% | — |