Other · model
LFM2.5-24B-A2B benchmark scores
As of 2026-10-29, LFM2.5-24B-A2B (Other) has measured scores on 12 benchmarks and estimated scores on 40 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 66.4% | estimated ± 3.0 pp, medium confidence |
| AA-HLE | 12.6% | estimated ± 3.8 pp, high confidence |
| AA-SciCode | 33.4% | estimated ± 3.0 pp, high confidence |
| CritPt | 0.4% | estimated ± 4.0 pp, high confidence |
| GDPval-AA | 11.7% | estimated ± 5.5 pp, medium confidence |
| GPQA Diamond | 74.4% | measured |
| HLE | 7.4% | measured |
| IFBench | 74.7% | measured |
| LiveCodeBench v6 | 71.7% | estimated ± 4.4 pp, high confidence |
| SciCode | 32.7% | measured |
| BrowseComp | 84.4% | estimated ± 7.0 pp, low confidence |
| IFEval | 96.8% | measured |
| τ²-bench results | 86.8% | estimated ± 12.4 pp, low confidence |
| AA Coding Index | 24.7% | estimated ± 5.3 pp, medium confidence |
| SWE-bench Verified | 51.0% | measured |
| AutomationBench | 27.7% | measured |
| BFCL v4 | 65.0% | measured |
| CyberGym | 79.0% | estimated ± 4.6 pp, medium confidence |
| SWE-bench Pro | 26.3% | estimated ± 4.6 pp, high confidence |
| τ³-bench results | 8.0% | measured |
| AA Agentic Index | 4.9% | estimated ± 5.9 pp, low confidence |
| MMLU-Pro | 78.3% | estimated ± 5.5 pp, medium confidence |
| AA-IFBench | 66.8% | estimated ± 7.4 pp, medium confidence |
| AIME 2024 | 74.3% | estimated ± 8.2 pp, medium confidence |
| Artificial Analysis Intelligence Index | 12.0% | estimated ± 6.3 pp, medium confidence |
| GPQA | 74.4% | measured |
| MMLU | 87.9% | estimated ± 4.3 pp, high confidence |
| AA-Omniscience Accuracy | 22.0% | estimated ± 8.1 pp, medium confidence |
| MATH-500 | 96.0% | measured |
| MMMLU | 79.8% | estimated ± 1.1 pp, low confidence |
| AA MMLU-Pro | 85.3% | estimated ± 1.0 pp, low confidence |
| AIME 2025 | 69.3% | estimated ± 14.2 pp, low confidence |
| AA Tau3 Banking | 41.8% | estimated ± 3.8 pp, medium confidence |
| ARC-AGI-1 | 29.3% | estimated ± 7.0 pp, low confidence |
| ARC-AGI-2 | 0.0% | estimated ± 11.6 pp, low confidence |
| Vals GPQA Diamond | 62.0% | estimated ± 4.8 pp, medium confidence |
| Vals LiveCodeBench | 82.1% | estimated ± 0.8 pp, low confidence |
| Vals MMLU-Pro | 44.5% | estimated ± 1.2 pp, medium confidence |
| Vals SWE-bench | 43.4% | estimated ± 5.3 pp, low confidence |
| CWE-Bench-AA | 55.7% | estimated ± 6.0 pp, low confidence |
| AA Cyber Index | 32.8% | estimated ± 9.4 pp, low confidence |
| HLE w/o tools | 10.5% | estimated ± 4.3 pp, medium confidence |
| Vibe Code Bench | 0.6% | estimated ± 8.1 pp, low confidence |
| C-Eval | 89.6% | estimated ± 1.2 pp, high confidence |
| MCP Atlas | 35.9% | estimated ± 14.2 pp, low confidence |
| MMLU-Redux | 91.6% | estimated ± 1.7 pp, high confidence |
| SuperGPQA | 50.2% | estimated ± 6.9 pp, medium confidence |
| MMLU-Pro (Arcee) | 80.9% | estimated ± 3.4 pp, medium confidence |
| React Native Evals | 71.4% | estimated ± 2.3 pp, low confidence |
| SimpleQA | 34.7% | estimated ± 14.6 pp, low confidence |
| LiveCodeBench | 59.0% | estimated ± 14.6 pp, low confidence |
| PinchBench | 83.2% | measured |