winner on BenchLM Math
OpenAI: GPT-5.597.4
direct benchmark result, not a broad vertical composite | source row dated 2026-05-28
scored on 2026-05-28
latest mapped results | top 20
| # | Model | Score | Evidence | Tested |
|---|---|---|---|---|
| 1 | OpenAI: GPT-5.5 | 97.4 | model-only aggregator | 2026-05-28 |
| 2 | Anthropic: Claude Opus 4.5 | 94.9 | model-only aggregator | 2026-05-28 |
| 3 | OpenAI: GPT-5.4 | 94.4 | model-only aggregator | 2026-05-28 |
| 4 | Z.ai: GLM 5 | 91.3 | model-only aggregator | 2026-05-28 |
| 5 | Anthropic: Claude Sonnet 4.5 | 87.7 | model-only aggregator | 2026-05-28 |
| 6 | Anthropic: Claude Opus 4.6 | 86.3 | model-only aggregator | 2026-05-28 |
| 7 | OpenAI: GPT-5.2 | 81.0 | model-only aggregator | 2026-05-28 |
| 8 | Anthropic: Claude Sonnet 4.6 | 75.7 | model-only aggregator | 2026-05-28 |
| 9 | Google: Gemini 2.5 Pro | 73.4 | model-only aggregator | 2026-05-28 |
| 10 | OpenAI: GPT-5.1 | 68.2 | model-only aggregator | 2026-05-28 |
what this result means
BenchLM composite math score (0-100).
This benchmark contributes direct public evidence. Read its scope before generalizing the result.
A win here is a win on BenchLM Math. Broad task pages require independent corroboration before naming a general winner.
source record
category: math
metric: score
matched models: 10
latest source date: 2026-05-28
direction: higher is better