GPT-5.5vsOpus 4.7
Across 15 shared benchmarks, GPT-5.5 leads overall: GPT-5.5 wins 10, Opus 4.7 wins 4, with 1 ties and an average score difference of +4.16.
GPT-5.510 wins(67%)Ties1(27%)4 winsOpus 4.7
Benchmark scores
Grouped by capability, sorted by largest gap within each. 15 shared benchmarks.
General Knowledge
GPT-5.5 3/6| Benchmark | GPT-5.5 | Opus 4.7 | Diff |
|---|---|---|---|
| ARC-AGI-2 | 852 / 62Thinking High (No Tools) | 75.8011 / 62最高（无工具） | +9.20 |
| LiveBench | 80.711 / 115Deep Thinking (No Tools) | 76.917 / 115Deep Thinking (No Tools) | +3.80 |
| HLE | 52.2020 / 172Thinking High (With Tools) | 54.7013 / 172Extended (with tools) | -2.50 |
| ARC-AGI | 955 / 68极高强度思考（无工具） | 93.5011 / 68Thinking High (No Tools) | +1.50 |
| GPQA Diamond | 93.606 / 187Thinking High (No Tools) | 94.204 / 187Extended (no tools) | -0.60 |
| ARC-AGI-3 | 05 / 9Thinking High (No Tools) | 08 / 9Thinking High (No Tools) | — |
AI Agent - Tool Usage
GPT-5.5 3/4| Benchmark | GPT-5.5 | Opus 4.7 | Diff |
|---|---|---|---|
| TerminalBench 2.1 | 83.407 / 27Thinking High (With Tools) | 69.7019 / 27Thinking High (With Tools) | +13.70 |
| Terminal Bench 2.0 | 82.701 / 47Thinking High (With Tools) | 69.406 / 47Extended (with tools) | +13.30 |
| MCP-Atlas | 75.3012 / 27极高强度思考（工具） | 79.107 / 27Deep Thinking (With Tools) | -3.80 |
| OSWorld-Verified | 78.708 / 24Thinking High (With Tools) | 7810 / 24Extended (with tools) | +0.70 |
Math and Reasoning
GPT-5.5 2/2| Benchmark | GPT-5.5 | Opus 4.7 | Diff |
|---|---|---|---|
| FrontierMath - Tier 4 | 35.407 / 80Thinking High (With Tools) | 22.9012 / 80极高强度思考（无工具） | +12.50 |
| FrontierMath | 51.702 / 60Thinking High (With Tools) | 43.806 / 60极高强度思考（无工具） | +7.90 |
AI Agent - Information Search
GPT-5.5 1/1| Benchmark | GPT-5.5 | Opus 4.7 | Diff |
|---|---|---|---|
| BrowseComp | 84.408 / 53Thinking High (With Tools + Internet) | 79.3017 / 53Extended (with tools) | +5.10 |
Coding and Software Engineer
Opus 4.7 1/1| Benchmark | GPT-5.5 | Opus 4.7 | Diff |
|---|---|---|---|
| SWE-Bench Pro - Public | 58.6013 / 54Thinking High (With Tools) | 64.307 / 54Extended (with tools) | -5.70 |
Commonsense Reasoning
GPT-5.5 1/1| Benchmark | GPT-5.5 | Opus 4.7 | Diff |
|---|---|---|---|
| Simple Bench | 697 / 63Normal (No Tools) | 61.7013 / 63Normal (No Tools) | +7.30 |
Specs
| Field | GPT-5.5 | 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 | Opus 4.7 |
|---|---|---|
| Text input | $0.5 / 1M tokens | $5 / 1M tokens |
| Text output | $30 / 1M tokens | $25 / 1M tokens |
| Cache read | $0.5 / 1M tokens | $0.5 / 1M tokens |
| Cache write | $6.25 / 1M tokens | $6.25 / 1M tokens |
Summary
- GPT-5.5leads in:General Knowledge (3/6), AI Agent - Tool Usage (3/4), Math and Reasoning (2/2), AI Agent - Information Search (1/1), Commonsense Reasoning (1/1)
- Opus 4.7leads in:Coding and Software Engineer (1/1)
On average across the 15 shared benchmarks, GPT-5.5 scores 4.16 higher.
Largest single-benchmark gap: TerminalBench 2.1 — GPT-5.5 83.40 vs Opus 4.7 69.70 (+13.70).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.