Model card
GPT-4.5.
OpenAIapiUndisclosed params
§ 01 · Benchmarks
Every benchmark GPT-4.5 has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | GSM8K | Reasoning · Mathematical Reasoning | accuracy | 98.2% | #10 | 2025-03-01 | source ↗ |
| 02 | SWE-Bench | Computer Code · Code Generation | resolve-rate-agentic | 62.0% | #16 | 2025-06-01 | |
| 03 | SWE-Bench | Computer Code · Code Generation | resolve-rate | 62.0% | #21 | 2025-06-01 | source ↗ |
| 04 | SWE-bench Verified | Agentic AI · SWE-bench | resolve-rate | 38.0% | #75 | — | source ↗ |
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.
§ 02 · Strengths by area
Where GPT-4.5 actually performs.
§ 03 · Papers
1 paper with results for GPT-4.5.
- 2023-10-10· Computer Code· 1 result
SWE-bench: Can Language Models Resolve Real-World GitHub Issues?
Carlos E. Jimenez, John Yang, Alexander Wettig, Shunyu Yao et al.
§ 04 · Related models
Other OpenAI models scored on Codesota.
§ 05 · Sources & freshness
Where these numbers come from.
gsm8k-shadow-page
1
result
codex
1
result
sota-timeline
1
result
editorial
1
result
2 of 4 rows marked verified. · first result 2025-03-01, latest 2025-06-01.