Claude Opus 5vsGPT-5.6 Sol
Across 8 shared benchmarks, Claude Opus 5 leads overall: Claude Opus 5 wins 6, GPT-5.6 Sol wins 1, with 1 ties and an average score difference of +39.40.
Claude Opus 5
Anthropic · 2026-07-24 · Reasoning model
GPT-5.6 Sol
OpenAI · 2026-06-26 · Reasoning model
Claude Opus 56 wins(75%)Ties1(13%)1 winGPT-5.6 Sol
Benchmark scores
Grouped by capability, sorted by largest gap within each. 8 shared benchmarks.
AI Agent - Tool Usage
Claude Opus 5 2/2| Benchmark | Claude Opus 5 | GPT-5.6 Sol | Diff |
|---|---|---|---|
| Terminal-Bench 4.0 | 51.823 / 13Max (With Tools) | 37.276 / 13Max (With Tools) | +14.55 |
| Terminal-Bench-Science 0.1 | 303 / 11Max (With Tools) | 22.404 / 11Max (With Tools) | +7.60 |
Productivity Knowledge
Claude Opus 5 2/2| Benchmark | Claude Opus 5 | GPT-5.6 Sol | Diff |
|---|---|---|---|
| GDPval-AA v2 | 1,8611 / 26Max (With Tools) | 1,7288 / 26Max (With Tools) | +133 |
| AA-AnalystAgent | 53.752 / 12Max (With Tools + Internet) | 47.505 / 12Max (With Tools + Internet) | +6.25 |
General Knowledge
Even 1/1| Benchmark | Claude Opus 5 | GPT-5.6 Sol | Diff |
|---|---|---|---|
| ARC-AGI-1 | 97.504 / 92Extra-High (No Tools) | 97.504 / 92Extra-High (No Tools) | — |
Math and Reasoning
GPT-5.6 Sol 1/1| Benchmark | Claude Opus 5 | GPT-5.6 Sol | Diff |
|---|---|---|---|
| FrontierMath v2 | 85.615 / 58Max (No Tools) | 89.121 / 58Max (No Tools) | -3.51 |
Text Embedding
Claude Opus 5 1/1| Benchmark | Claude Opus 5 | GPT-5.6 Sol | Diff |
|---|---|---|---|
| Context Arena | 97.721 / 126Max (No Tools) | 97.632 / 126Max (No Tools) | +0.09 |
Writing and Creative Capabilities
Claude Opus 5 1/1| Benchmark | Claude Opus 5 | GPT-5.6 Sol | Diff |
|---|---|---|---|
| Creative Writing | 2,1213 / 106Normal (No Tools) | 1,9636 / 106Normal (No Tools) | +157.20 |
Specs
| Field | Claude Opus 5 | GPT-5.6 Sol |
|---|---|---|
| Publisher | Anthropic | OpenAI |
| Release date | 2026-07-24 | 2026-06-26 |
| Model type | Reasoning model | Reasoning model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1M | 1.05M |
| Max output | 128K | 128K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Claude Opus 5 | GPT-5.6 Sol |
|---|---|---|
| Text input | $5 / 1M tokens | $4 / 1M tokens |
| Text output | $25 / 1M tokens | $20 / 1M tokens |
| Cache read | $0.5 / 1M tokens | $0.4 / 1M tokens |
| Cache write | $6.25 / 1M tokens | $5 / 1M tokens |
Summary
- Claude Opus 5leads in:AI Agent - Tool Usage (2/2), Productivity Knowledge (2/2), Text Embedding (1/1), Writing and Creative Capabilities (1/1)
- GPT-5.6 Solleads in:Math and Reasoning (1/1)
- Tied in:General Knowledge
On average across the 8 shared benchmarks, Claude Opus 5 scores 39.40 higher.
Largest single-benchmark gap: Creative Writing — Claude Opus 5 2,121 vs GPT-5.6 Sol 1,963 (+157.20).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.