Model card
GPT-5.
OpenAIproprietary
§ 02 · Benchmarks
Every benchmark GPT-5 has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | GSM8K | Reasoning · Mathematical Reasoning | accuracy | 99.2% | #3 | 2025-08-01 | source ↗ |
| 02 | LiveCodeBench | Computer Code · Code Generation | pass@1 | 85.0% | #3 | — | source ↗ |
| 03 | LiveCodeBench Pro | Computer Code · Code Generation | elo | 2176.00 | #3 | — | source ↗ |
| 04 | HumanEval | Computer Code · Code Generation | pass@1 | 95.1% | #4 | 2025-12-01 | source ↗ |
| 05 | GPQA Diamond | Reasoning · Multi-step Reasoning | accuracy | 89.0% | #7 | — | source ↗ |
| 06 | MMLU | Reasoning · Commonsense Reasoning | accuracy | 90.8% | #8 | 2025-09-01 | source ↗ |
| 07 | SWE-Bench Verified | Computer Code · Code Generation | resolve-rate | 74.9% | #13 | — | source ↗ |
| 08 | SWE-bench Verified | Agentic AI · SWE-bench | resolve-rate | 74.9% | #20 | — | source ↗ |
| 09 | HLE | Reasoning · Multi-step Reasoning | accuracy | 25.3% | #22 | — | source ↗ |
| 10 | HLE | Reasoning · Multi-step Reasoning | accuracy | 25.3% | #23 | — |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area
Where GPT-5 actually performs.
§ 05 · Related models
Other OpenAI models scored on Codesota.
§ 06 · Sources & freshness
Where these numbers come from.
editorial
2
results
gsm8k-shadow-page-timeline
1
result
artificial-analysis
1
result
livecodebench-pro-official
1
result
shadow-page-humaneval
1
result
openai-gpt-5-launch
1
result
codesota-shadow-mmlu
1
result
openai-blog
1
result
scale-hle-official
1
result
4 of 10 rows marked verified. · first result 2025-08-01, latest 2025-12-01.