mapika: decider-2b-v8
1 benchmarks
Leaderboard position
#608
Overall rank (of 1121 models)
847
Elo rating
1
Benchmarks
Top 54.2%
Percentile
In mapika (of 3 models)
Above: Google: gemini-2.5-flash-native-audio #607
Below: Meta Llama: Llama 4 Scout Instruct #609
Benchmark highlights
Best benchmark
OpenJev (%)
64.1
Worst benchmark
OpenJev (%)
64.1
Nearest competitor from another company
The model from another company with the most similar benchmark scores.
No alternative found.
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 | ~86.26 | 50% | — |
| AIME 2025 | ~65.86 | 50% | — |
| AIME 2026 | ~73.35 | 50% | — |
| HMMT Feb 2026 | ~64.86 | 50% | — |
| GPQA Diamond | ~57.92 | 50% | — |
| MMLU-Pro | ~61.04 | 50% | — |
| Humanity's Last Exam (no tools) | ~11.5 | 50% | — |
| IFEval | ~82.17 | 50% | — |
| IFBench | ~53.21 | 50% | — |
| HumanEval | ~73.49 | 50% | — |
| MBPP | ~76.76 | 50% | — |
| Terminal-Bench v2.1 | ~21.25 | 50% | — |
| MMLU | ~67.48 | 50% | — |
| SWE-bench Verified | ~38.96 | 50% | — |
| SWE-bench Pro | ~17.92 | 50% | — |
| SciCode | ~26.81 | 50% | — |
| AA-LCR | ~42.95 | 50% | — |
| Artificial Analysis Intelligence Index v5.0.1 | ~12 | 50% | — |
| Humanity's Last Exam (with tools) | ~15.68 | 50% | — |
| GSM8K | ~82.54 | 50% | — |
| -(CoderPPL) | ~-1.69 | 50% | — |
| LiveCodeBench v6 | ~55.76 | 50% | — |
| MMMU-Pro | ~52.81 | 50% | — |
| MATH | ~60.45 | 50% | — |
| BBH | ~65.17 | 50% | — |