GPT-5.5vsGPT-5.2
Across 10 shared benchmarks, GPT-5.5 leads overall: GPT-5.5 wins 9, GPT-5.2 wins 1, with 0 ties and an average score difference of +178.94.
GPT-5.59 wins(90%)(10%)1 winGPT-5.2
Benchmark scores
Grouped by capability, sorted by largest gap within each. 10 shared benchmarks.
General Knowledge
GPT-5.5 4/4| Benchmark | GPT-5.5 | GPT-5.2 | Diff |
|---|---|---|---|
| ARC-AGI-2 | 851 / 59Thinking High (No Tools) | 54.2020 / 59深度思考（无工具、并行） | +30.80 |
| HLE | 52.2013 / 157Thinking High (With Tools) | 45.5032 / 157Deep Thinking (With Tools + Internet) | +6.70 |
| ARC-AGI | 953 / 65极高强度思考（无工具） | 90.5015 / 65深度思考（无工具、并行） | +4.50 |
| GPQA Diamond | 93.606 / 178Thinking High (No Tools) | 93.208 / 178深度思考（无工具、并行） | +0.40 |
Math and Reasoning
GPT-5.5 2/2| Benchmark | GPT-5.5 | GPT-5.2 | Diff |
|---|---|---|---|
| FrontierMath - Tier 4 | 35.407 / 80Thinking High (With Tools) | 18.8016 / 80Thinking High (No Tools) | +16.60 |
| FrontierMath | 51.702 / 60Thinking High (With Tools) | 40.308 / 60极高强度思考（工具） | +11.40 |
Agent Level Benchmark
GPT-5.2 1/1| Benchmark | GPT-5.5 | GPT-5.2 | Diff |
|---|---|---|---|
| τ²-Bench - Telecom | 985 / 35Thinking High (With Tools) | 98.704 / 35极高强度思考（工具） | -0.70 |
AI Agent - Information Search
GPT-5.5 1/1| Benchmark | GPT-5.5 | GPT-5.2 | Diff |
|---|---|---|---|
| BrowseComp | 84.405 / 45Thinking High (With Tools + Internet) | 65.8024 / 45Deep Thinking (With Tools + Internet) | +18.60 |
Coding and Software Engineer
GPT-5.5 1/1| Benchmark | GPT-5.5 | GPT-5.2 | Diff |
|---|---|---|---|
| SWE-Bench Pro - Public | 58.607 / 43Thinking High (With Tools) | 55.6017 / 43极高强度思考（工具） | +3 |
Productivity Knowledge
GPT-5.5 1/1| Benchmark | GPT-5.5 | GPT-5.2 | Diff |
|---|---|---|---|
| GDPval-AA | 1,7692 / 21Thinking High (No Tools) | 70.909 / 21Thinking High (With Tools) | +1,698 |
Specs
| Field | GPT-5.5 | GPT-5.2 |
|---|---|---|
| Publisher | OpenAI | OpenAI |
| Release date | 2026-04-23 | 2025-12-11 |
| Model type | Reasoning model | Chat model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1000K | 400K |
| Max output | 128K | Not available |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | GPT-5.5 | GPT-5.2 |
|---|---|---|
| Text input | $0.5 / 1M tokens | $1.75 / 1M tokens |
| Text output | $30 / 1M tokens | $14 / 1M tokens |
| Cache read | $0.5 / 1M tokens | $0.175 / 1M tokens |
| Cache write | $6.25 / 1M tokens | $1.75 / 1M tokens |
Summary
- GPT-5.5leads in:General Knowledge (4/4), Math and Reasoning (2/2), AI Agent - Information Search (1/1), Coding and Software Engineer (1/1), Productivity Knowledge (1/1)
- GPT-5.2leads in:Agent Level Benchmark (1/1)
On average across the 10 shared benchmarks, GPT-5.5 scores 178.94 higher.
Largest single-benchmark gap: GDPval-AA — GPT-5.5 1,769 vs GPT-5.2 70.90 (+1,698).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.