Other · model
LFM2.5-VL-450M benchmark scores
As of 2026-11-22, LFM2.5-VL-450M (Other) has measured scores on 7 benchmarks and estimated scores on 34 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 23.3% | estimated ± 6.2 pp, medium confidence |
| AA-HLE | 0.0% | estimated ± 5.7 pp, medium confidence |
| Artificial Analysis Intelligence Index | 9.4% | estimated ± 7.4 pp, low confidence |
| CritPt | 0.3% | estimated ± 4.0 pp, high confidence |
| GPQA Diamond | 26.3% | estimated ± 1.2 pp, high confidence |
| IFBench | 12.6% | estimated ± 9.2 pp, low confidence |
| AA-Omniscience Accuracy | 19.0% | estimated ± 8.5 pp, low confidence |
| IFEval | 61.2% | measured |
| τ²-bench results | 39.8% | estimated ± 12.4 pp, low confidence |
| AA-MMMU-Pro | 47.7% | estimated ± 3.0 pp, medium confidence |
| BFCL v4 | 21.1% | measured |
| MMLU-Pro | 19.3% | measured |
| AA-IFBench | 10.7% | estimated ± 8.8 pp, medium confidence |
| BBH | 42.0% | estimated ± 2.8 pp, high confidence |
| GPQA | 25.7% | measured |
| MMLU | 44.2% | estimated ± 2.1 pp, high confidence |
| Vals GPQA Diamond | 24.0% | estimated ± 4.5 pp, medium confidence |
| Vals MMLU-Pro | 46.9% | estimated ± 2.3 pp, medium confidence |
| MMMLU | 51.1% | estimated ± 1.7 pp, medium confidence |
| MMMU | 32.7% | measured |
| AI2D_TEST | 87.6% | estimated ± 1.4 pp, low confidence |
| C-Eval | 65.1% | estimated ± 1.2 pp, medium confidence |
| CountBench | 73.3% | measured |
| DynaMath | 73.4% | estimated ± 5.8 pp, low confidence |
| ERQA | 47.1% | estimated ± 5.2 pp, low confidence |
| MathVision | 25.2% | estimated ± 3.6 pp, medium confidence |
| MLVU (M-Avg) | 39.2% | estimated ± 2.4 pp, medium confidence |
| MMLU-Redux | 58.0% | estimated ± 0.6 pp, medium confidence |
| MMMU-Pro | 20.5% | estimated ± 2.2 pp, medium confidence |
| RealWorldQA | 58.4% | measured |
| ScreenSpot Pro | 55.3% | estimated ± 7.9 pp, low confidence |
| SimpleVQA | 21.6% | estimated ± 7.1 pp, low confidence |
| SuperGPQA | 7.0% | estimated ± 12.4 pp, low confidence |
| Video-MME (w/o subtitle) | 58.0% | estimated ± 1.2 pp, medium confidence |
| Video-MME (with subtitle) | 48.0% | estimated ± 3.2 pp, medium confidence |
| VideoMMMU | 38.7% | estimated ± 2.3 pp, medium confidence |
| We-Math | 53.7% | estimated ± 7.2 pp, low confidence |
| SimpleQA | 14.6% | estimated ± 13.4 pp, low confidence |
| MMVU | 37.8% | estimated ± 3.8 pp, low confidence |
| RefCOCO (avg) | 81.9% | estimated ± 3.1 pp, low confidence |
| OCRBench V2 | 31.9% | estimated ± 5.4 pp, low confidence |