Claude Opus 5.5 vs GPT-5.2-Codex
Claude Opus 5.5 
GPT-5.2-Codex Verdict
- Claude Opus 5.5 (thinking) leads on quality: 70.7 vs 60.3.
- Claude Opus 5.5 (thinking) is stronger in composite, knowledge, long context, multimodal, reasoning.
- GPT-5.2-Codex (xhigh) is stronger in agents & tools, instruction following.
- GPT-5.2-Codex (xhigh) is 1.7× cheaper ($4.81 vs $8.00 per 1M blended).
- Claude Opus 5.5 (thinking) streams 1.7× faster (48 vs 28 tokens per second).
| Metric | Claude Opus 5.5 (thinking) | GPT-5.2-Codex (xhigh) |
|---|---|---|
| BenchLeader Index | 70.7 | 60.3 |
| Composite score | 95.0 | 64.0 |
| Knowledge score | 85.0 | 62.3 |
| Long context score | 68.5 | 67.3 |
| Multimodal score | 71.9 | 60.2 |
| Reasoning score | 95.0 | 57.6 |
| Agents & tools score | – | 65.9 |
| Instruction following score | – | 75.3 |
| Blended price $/M | $8.00 | $4.81 |
| Output speed | 48 tok/s | 28 tok/s |
| Time to first answer | 4.2 s | 3.6 s |
| Context window | 1M | 400k |
| AA Intelligence Index | 57.6 | 28.5 |
| IFBench | – | 77.6% |
| AA-LCR | 84.7% | 82.3% |
| MMMU-Pro | 87.7% | 76.3% |
| AA-Omniscience | 46.4 | -2.2 |
| Terminal-Bench Hard | – | 37.1% |
| GPQA Diamond (AA) | – | 89.9% |
| Humanity's Last Exam (AA) | 61.4% | 35.7% |
| SciCode (AA) | 66.9% | – |
| τ²-Bench Telecom (AA) | – | 92.1% |
| CritPt | 31.7% | 8.7% |
| GDPval (AA) | 67.3% | – |
| Terminal-Bench 4.0 (AA) | 59.6% | – |
| AutomationBench | 69.5% | – |
| GDP.pdf | 26.2% | – |
| Harvey LAB | 91.2% | – |
| AA-Omniscience: accuracy | 66.2% | 41.1% |
| AA-Omniscience: non-hallucination | 41.4% | 26.6% |
| 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-5.2-Codex: questions
- Is Claude Opus 5.5 better than GPT-5.2-Codex?
- Claude Opus 5.5 (thinking) leads on quality: 70.7 vs 60.3. 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-5.2-Codex?
- GPT-5.2-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-5.2-Codex?
- Claude Opus 5.5 streams faster: 48 against 28 output tokens per second.
- Which has the larger context window?
- Claude Opus 5.5 accepts more context: 1M against 400k tokens.