winner on BenchLM Math
OpenAI: GPT-5.593.8
direct benchmark result, not a broad vertical composite | source row dated 2026-07-09
scored on 2026-07-09
latest mapped results | top 20
| # | Model | Score | Evidence | Tested |
|---|---|---|---|---|
| 1 | OpenAI: GPT-5.5 | 93.8 | model-only aggregator | 2026-07-09 |
| 2 | OpenAI: GPT-5.4 | 87.0 | model-only aggregator | 2026-07-09 |
| 3 | DeepSeek: DeepSeek V4 Flash | 85.4 | model-only aggregator | 2026-07-09 |
| 4 | Anthropic: Claude Sonnet 4.5 | 80.0 | model-only aggregator | 2026-07-06 |
| 5 | Anthropic: Claude Opus 4.5 | 78.4 | model-only aggregator | 2026-07-09 |
| 6 | Z.ai: GLM 5 | 68.9 | model-only aggregator | 2026-07-09 |
| 7 | Google: Gemini 2.5 Pro | 68.1 | model-only aggregator | 2026-07-06 |
| 8 | OpenAI: GPT-5.2 | 65.1 | model-only aggregator | 2026-07-09 |
| 9 | Anthropic: Claude Opus 4.6 | 61.6 | model-only aggregator | 2026-07-09 |
| 10 | Anthropic: Claude Sonnet 4.6 | 48.6 | model-only aggregator | 2026-07-09 |
| 11 | OpenAI: GPT-5.1 | 44.1 | model-only aggregator | 2026-07-09 |
| 12 | DeepSeek: DeepSeek V4 Pro | 28.5 | model-only aggregator | 2026-07-09 |
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: 12
latest source date: 2026-07-09
direction: higher is better