nullsilver: alpha-sys-1-3b
1 benchmarks
Leaderboard position
#557
Overall rank (of 1121 models)
980
Elo rating
1
Benchmarks
Top 49.7%
Percentile
In nullsilver (of 3 models)
Above: NVIDIA Corporation: nvidia-nemotron-3-nano-30b-a3b #556
Below: inclusionAI: ling-1t #558
Benchmark highlights
Best benchmark
CIFAR-10
93.8
Worst benchmark
CIFAR-10
93.8
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 | ~88.56 | 50% | — |
| AIME 2025 | ~70.22 | 50% | — |
| AIME 2026 | ~75.92 | 50% | — |
| HMMT Feb 2026 | ~67.44 | 50% | — |
| GPQA Diamond | ~60.85 | 50% | — |
| MMLU-Pro | ~64.3 | 50% | — |
| Humanity's Last Exam (no tools) | ~13.87 | 50% | — |
| IFEval | ~83.99 | 50% | — |
| IFBench | ~55.41 | 50% | — |
| HumanEval | ~77.36 | 50% | — |
| MBPP | ~81.81 | 50% | — |
| Terminal-Bench v2.1 | ~25.86 | 50% | — |
| MMLU | ~70.91 | 50% | — |
| SWE-bench Verified | ~43.13 | 50% | — |
| SWE-bench Pro | ~21.81 | 50% | — |
| SciCode | ~28.88 | 50% | — |
| AA-LCR | ~45.55 | 50% | — |
| Artificial Analysis Intelligence Index v5.0.1 | ~12 | 50% | — |
| Humanity's Last Exam (with tools) | ~18.84 | 50% | — |
| GSM8K | ~87.54 | 50% | — |
| -(CoderPPL) | ~-1.69 | 50% | — |
| LiveCodeBench v6 | ~58.8 | 50% | — |
| MMMU-Pro | ~55.31 | 50% | — |
| MATH | ~66.03 | 50% | — |
| BBH | ~68.12 | 50% | — |