GPT-5.5 ProvsOpus 4.7
Across 10 shared benchmarks, GPT-5.5 Pro leads overall: GPT-5.5 Pro wins 9, Opus 4.7 wins 1, with 0 ties and an average score difference of +12.80.
GPT-5.5 Pro
OpenAI · 2026-04-23 · Reasoning model
Opus 4.7
Anthropic · 2026-04-16 · Reasoning model
GPT-5.5 Pro9 wins(90%)(10%)1 winOpus 4.7
Benchmark scores
Grouped by capability, sorted by largest gap within each. 10 shared benchmarks.
Math and Reasoning
GPT-5.5 Pro 4/4| Benchmark | GPT-5.5 Pro | Opus 4.7 | Diff |
|---|---|---|---|
| FrontierMath Tier 4 v2 | 78.053 / 34极高强度思考（无工具） | 31.7114 / 34最高（无工具） | +46.34 |
| FrontierMath v2 | 87.722 / 34极高强度思考（无工具） | 70.1814 / 34最高（无工具） | +17.54 |
| FrontierMath - Tier 4 | 39.601 / 80Thinking High (No Tools) | 22.9012 / 80极高强度思考（无工具） | +16.70 |
| FrontierMath | 52.401 / 60极高强度思考（工具） | 43.806 / 60极高强度思考（无工具） | +8.60 |
General Knowledge
GPT-5.5 Pro 3/3| Benchmark | GPT-5.5 Pro | Opus 4.7 | Diff |
|---|---|---|---|
| ARC-AGI-2 | 84.604 / 62Thinking High (No Tools) | 75.8011 / 62最高（无工具） | +8.80 |
| ARC-AGI | 96.502 / 68Thinking High (No Tools) | 93.5011 / 68Thinking High (No Tools) | +3 |
| HLE | 57.2010 / 181极高强度思考（工具） | 54.7015 / 181Extended (with tools) | +2.50 |
AI Agent - Information Search
GPT-5.5 Pro 1/1| Benchmark | GPT-5.5 Pro | Opus 4.7 | Diff |
|---|---|---|---|
| BrowseComp | 90.103 / 54Deep Thinking (With Tools + Internet) | 79.3017 / 54Extended (with tools) | +10.80 |
Commonsense Reasoning
GPT-5.5 Pro 1/1| Benchmark | GPT-5.5 Pro | Opus 4.7 | Diff |
|---|---|---|---|
| SimpleBench | 76.903 / 67Normal (No Tools) | 62.9012 / 67Normal (No Tools) | +14 |
General Evaluation
Opus 4.7 1/1| Benchmark | GPT-5.5 Pro | Opus 4.7 | Diff |
|---|---|---|---|
| GPQA Diamond | 93.928 / 226极高强度思考（无工具） | 94.205 / 226Extended (no tools) | -0.28 |
Specs
| Field | GPT-5.5 Pro | Opus 4.7 |
|---|---|---|
| Publisher | OpenAI | Anthropic |
| Release date | 2026-04-23 | 2026-04-16 |
| Model type | Reasoning model | Reasoning model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1000K | 1000K |
| Max output | 128K | 128K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | GPT-5.5 Pro | Opus 4.7 |
|---|---|---|
| Text input | $30 / 1M tokens | $5 / 1M tokens |
| Text output | $180 / 1M tokens | $25 / 1M tokens |
| Cache read | Not public | $0.5 / 1M tokens |
| Cache write | Not public | $6.25 / 1M tokens |
Summary
- GPT-5.5 Proleads in:Math and Reasoning (4/4), General Knowledge (3/3), AI Agent - Information Search (1/1), Commonsense Reasoning (1/1)
- Opus 4.7leads in:General Evaluation (1/1)
On average across the 10 shared benchmarks, GPT-5.5 Pro scores 12.80 higher.
Largest single-benchmark gap: FrontierMath Tier 4 v2 — GPT-5.5 Pro 78.05 vs Opus 4.7 31.71 (+46.34).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.