Claude Opus 5.5 vs GPT-5.6 Sol
MMMU-Pro (vision) MMMU-Pro (vision): College-level questions that require understanding images: charts, diagrams, photos
Opus 5.585.7
GPT-5.6 Sol81.4
Chartography Chartography: Expert chart reasoning over professional scientific, medical, financial and engineering charts
Opus 5.5 · max66.3
GPT-5.6 Sol · max45
GDP.pdf · AA all-pass GDP.pdf · AA all-pass: Models reason over professional PDF documents; an attempt passes only when every atomic criterion passes
Opus 5.525.6
GPT-5.6 Sol26.2
Arena Vision · overall Arena Vision · overall: Human pairwise preferences in the stated vision vote pool
Opus 5.5 · high1,320.8
GPT-5.6 Sol · xhigh1,280.6
Arena Document · overall Arena Document · overall: Human pairwise preferences in the stated document vote pool
Opus 5.5—
GPT-5.6 Sol · xhigh1,482.7
| Benchmark | Opus 5.5 | GPT-5.6 Sol |
|---|---|---|
| MMMU-Pro (vision) | 85.7 | 81.4 |
| Chartography | 66.3 · max | 45 · max |
| GDP.pdf · AA all-pass | 25.6 | 26.2 |
| Arena Vision · overall | 1,320.8 · high | 1,280.6 · xhigh |
| Arena Document · overall | — | 1,482.7 · xhigh |
Specs
| Spec | Opus 5.5 | GPT-5.6 Sol |
|---|---|---|
| Input, per 1M | $4 | $4 |
| Output, per 1M | $20 | $20 |
| Cost per task? | $1.34 | $0.50 |
| Speed, tok/s | 81 | 85 |
| Context | 1M | 1.05M |
| Weights | Closed | Closed |
| Released | Sep 2026 | Jul 2026 |
| See images | ✓ | ✓ |
| Read PDFs | ✓ | Partial |
| Hear audio | — | — |
| Watch video | — | — |
| Call tools | ✓ | ✓ |
| Force a tool call | — | Partial |
| Return strict JSON | ✓ | ✓ |
| Search the web | ✓ | ? |
| Run code | ✓ | ? |
| Use a computer | ✓ | ? |
What breaks when you switch
Opus 5.5 → GPT-5.6 Sol
Nothing known.
GPT-5.6 Sol → Opus 5.5
Context: 1.05M → 1M
No forced tool use›
Forced tool_choice any or tool returns HTTP 400; remove the forced choice.
Thinking is always on›
Disabled thinking or a manual budget returns HTTP 400; omit thinking or use adaptive thinking.
Remove sampling parameters›
Non-default temperature, top_p or top_k returns HTTP 400; omit these parameters and use prompting.
No assistant prefill›
Assistant prefill returns HTTP 400; use structured outputs or system instructions, with system instructions on Bedrock.
Old computer tool rejected · Claude API / Google Cloud›
The computer_20251124 tool returns HTTP 400 on the Claude API and Google Cloud; use computer_toolset_20260801 there.
Return thinking blocks unmodified›
Edited, reordered or partially dropped thinking blocks return HTTP 400; echo assistant messages complete and unmodified with tool results.