Composer 2: benchmark results
Provider: Other.
Aggregate Rating 1680 ± 30, rank #447 of 10916 rated models, from 13 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SvelteBench | 95.6 | Average pass@1 (%) | 77.1 |
| SlopCodeBench (Just Solve) - Core | 52.6 | Core solve rate (%): share of the 196 checkpoints whose work | 47.1 |
| SlopCodeBench (Just Solve) - Isolated | 16.3 | Isolated solve rate (%): share of the 196 checkpoints whose | 47.1 |
| SlopCodeBench (Just Solve) | 6.1 | Strict solve rate (%): share of the 196 checkpoints whose wo | 41.2 |
| LLM Chess (Saplin) | -63.07 | Native anchored-opponent Elo (signed points) | 37 |
| LM Market Cap LMC Score | 65.7 | LMC Score (0-100) | 30.1 |
| AgentKernelArena - PyTorch-to-HIP | 4.14 | Mean speedup (x; arithmetic mean over the 24 PyTorch-to-HIP | 28.6 |
| Backtrader-Bench - Mined Questions (No Tools) | 23.7 | Accuracy (%; one run on the 38 four-option questions mined b | 22.2 |
| Backtrader-Bench (No Tools) | 51 | Accuracy (%; mean of 10 runs on the curated set of 30 four-o | 11.1 |
| AgentKernelArena - HIP-to-HIP | 1.44 | Mean speedup (x; arithmetic mean over the 24 HIP-to-HIP task | 0 |
| AgentKernelArena - Triton-to-Triton | 1.59 | Mean speedup (x; arithmetic mean over the 148 Triton-to-Trit | 0 |
| Backtrader-Bench (With Tools) | 76.7 | Accuracy (%; one pass over the curated set of 30 four-option | 0 |
Interactive version: theaggregate.ai/model?slug=composer-2 · How It Works · Data refreshed daily, snapshot 2026-11-08.