GPT-5.6 SolvsClaude Opus 4.8
Across 13 shared benchmarks, GPT-5.6 Sol leads overall: GPT-5.6 Sol wins 10, Claude Opus 4.8 wins 3, with 0 ties and an average score difference of +23.04.
GPT-5.6 Sol
OpenAI · 2026-06-26 · Reasoning model
Claude Opus 4.8
Anthropic · 2026-05-28 · Reasoning model
GPT-5.6 Sol10 wins(77%)(23%)3 winsClaude Opus 4.8
Benchmark scores
Grouped by capability, sorted by largest gap within each. 13 shared benchmarks.
Coding and Software Engineer
GPT-5.6 Sol 3/4| Benchmark | GPT-5.6 Sol | Claude Opus 4.8 | Diff |
|---|---|---|---|
| Text Arena (Coding) | 1,6204 / 35极高强度思考（无工具） | 1,5459 / 35Normal (No Tools) | +75.22 |
| WeirdML v2 | 88.762 / 52Thinking High (With Tools) | 70.4518 / 52Normal (With Tools) | +18.31 |
| DeepSWE | 72.701 / 32极高强度思考（工具） | 5917 / 32Deep Thinking (With Tools) | +13.70 |
| SWE-Bench Pro - Public | 64.608 / 59极高强度思考（工具） | 69.204 / 59Extended (with tools) | -4.60 |
AI Agent - Tool Usage
GPT-5.6 Sol 3/3| Benchmark | GPT-5.6 Sol | Claude Opus 4.8 | Diff |
|---|---|---|---|
| Terminal-Bench 4.0 | 37.275 / 11Max (With Tools) | 23.646 / 11Max (With Tools) | +13.63 |
| Terminal-Bench-Science 0.1 | 22.403 / 10Max (With Tools) | 10.505 / 10Max (With Tools) | +11.90 |
| Terminal-Bench 2.1 | 88.801 / 47最高（无工具） | 78.9023 / 47Thinking High (With Tools) | +9.90 |
Math and Reasoning
GPT-5.6 Sol 2/2| Benchmark | GPT-5.6 Sol | Claude Opus 4.8 | Diff |
|---|---|---|---|
| FrontierMath Tier 4 v2 | 82.932 / 40最高（无工具） | 56.109 / 40最高（无工具） | +26.83 |
| FrontierMath v2 | 89.121 / 58最高（无工具） | 809 / 58最高（无工具） | +9.12 |
General Evaluation
Claude Opus 4.8 1/1| Benchmark | GPT-5.6 Sol | Claude Opus 4.8 | Diff |
|---|---|---|---|
| GPQA Diamond | 82.83121 / 270Normal (No Tools) | 85.3596 / 270Normal (No Tools) | -2.53 |
General Knowledge
Claude Opus 4.8 1/1| Benchmark | GPT-5.6 Sol | Claude Opus 4.8 | Diff |
|---|---|---|---|
| HLE | 49.5038 / 188最高（无工具） | 57.9010 / 188Extended (with tools) | -8.40 |
Text Embedding
GPT-5.6 Sol 1/1| Benchmark | GPT-5.6 Sol | Claude Opus 4.8 | Diff |
|---|---|---|---|
| Context Arena | 97.632 / 126最高（无工具） | 90.0415 / 126最高（无工具） | +7.59 |
Writing and Creative Capabilities
GPT-5.6 Sol 1/1| Benchmark | GPT-5.6 Sol | Claude Opus 4.8 | Diff |
|---|---|---|---|
| Creative Writing | 1,9644 / 99Normal (No Tools) | 1,83513 / 99Normal (No Tools) | +128.80 |
Specs
| Field | GPT-5.6 Sol | Claude Opus 4.8 |
|---|---|---|
| Publisher | OpenAI | Anthropic |
| Release date | 2026-06-26 | 2026-05-28 |
| Model type | Reasoning model | Reasoning model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1.05M | 1M |
| Max output | 128K | 125K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | GPT-5.6 Sol | Claude Opus 4.8 |
|---|---|---|
| Text input | $4 / 1M tokens | $5 / 1M tokens |
| Text output | $20 / 1M tokens | $25 / 1M tokens |
| Cache read | $0.4 / 1M tokens | $0.5 / 1M tokens |
| Cache write | $5 / 1M tokens | $6.25 / 1M tokens |
Summary
- GPT-5.6 Solleads in:Coding and Software Engineer (3/4), AI Agent - Tool Usage (3/3), Math and Reasoning (2/2), Text Embedding (1/1), Writing and Creative Capabilities (1/1)
- Claude Opus 4.8leads in:General Evaluation (1/1), General Knowledge (1/1)
On average across the 13 shared benchmarks, GPT-5.6 Sol scores 23.04 higher.
Largest single-benchmark gap: Creative Writing — GPT-5.6 Sol 1,964 vs Claude Opus 4.8 1,835 (+128.80).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.