Claude Opus 5.5 vs GPT-5.3-Codex
Claude Opus 5.5 
GPT-5.3-Codex Verdict
- Claude Opus 5.5 (thinking) leads on quality: 70.7 vs 64.9.
- Claude Opus 5.5 (thinking) is stronger in composite, knowledge, long context, multimodal, reasoning.
- GPT-5.3-Codex (xhigh) is stronger in agents & tools, coding, instruction following.
- GPT-5.3-Codex (xhigh) is 1.7× cheaper ($4.81 vs $8.00 per 1M blended).
- GPT-5.3-Codex (xhigh) streams 3.3× faster (157 vs 48 tokens per second).
| Metric | Claude Opus 5.5 (thinking) | GPT-5.3-Codex (xhigh) |
|---|---|---|
| BenchLeader Index | 70.7 | 64.9 |
| Composite score | 95.0 | 68.8 |
| Knowledge score | 85.0 | 68.4 |
| Long context score | 68.5 | 67.8 |
| Multimodal score | 71.9 | 62.5 |
| Reasoning score | 95.0 | 72.6 |
| Agents & tools score | – | 79.8 |
| Coding score | – | 61.5 |
| Instruction following score | – | 73.4 |
| Blended price $/M | $8.00 | $4.81 |
| Output speed | 48 tok/s | 157 tok/s |
| Time to first answer | 4.2 s | 57.5 s |
| Context window | 1M | 400k |
| WeirdML | – | 77.9% |
| AA Intelligence Index | 57.6 | 32.5 |
| IFBench | – | 75.4% |
| AA-LCR | 84.7% | 83.3% |
| MMMU-Pro | 87.7% | 78.5% |
| AA-Omniscience | 46.4 | 10.9 |
| Terminal-Bench Hard | – | 53.0% |
| GPQA Diamond (AA) | – | 91.5% |
| Humanity's Last Exam (AA) | 61.4% | 42.5% |
| SciCode (AA) | 66.9% | – |
| τ²-Bench Telecom (AA) | – | 86.0% |
| LiveCodeBench | – | 87.3% |
| IOI | – | 53.8% |
| SWE-bench (Vals) | – | 78.0% |
| CritPt | 31.7% | 16.9% |
| GDPval (AA) | 67.3% | – |
| Vibe Code Bench v1.1 | – | 61.8% |
| ALE-Bench | – | 1655.2 |
| Terminal-Bench 4.0 (AA) | 59.6% | – |
| AutomationBench | 69.5% | – |
| GDP.pdf | 26.2% | – |
| Harvey LAB | 91.2% | – |
| AA-Omniscience: accuracy | 66.2% | 52.9% |
| AA-Omniscience: non-hallucination | 41.4% | 10.8% |
| AA-Briefcase | 1822 | – |
Data as of 2026-09-22. Best configuration of each model; every score links to its source on the model pages.
Claude Opus 5.5 vs GPT-5.3-Codex: questions
- Is Claude Opus 5.5 better than GPT-5.3-Codex?
- Claude Opus 5.5 (thinking) leads on quality: 70.7 vs 64.9. The BenchLeader Index combines every independent quality benchmark; Claude Opus 5.5 (thinking) is ahead overall as of 2026-09-22, but check the category scores for your use.
- Which is cheaper, Claude Opus 5.5 or GPT-5.3-Codex?
- GPT-5.3-Codex is cheaper: $4.81 against $8.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Claude Opus 5.5 or GPT-5.3-Codex?
- GPT-5.3-Codex streams faster: 157 against 48 output tokens per second.
- Which has the larger context window?
- Claude Opus 5.5 accepts more context: 1M against 400k tokens.