Claude Opus 4.5 vs GPT-5.2
Claude Opus 4.5 
GPT-5.2 Verdict
- Claude Opus 4.5 (thinking) and GPT-5.2 are level on quality (60.7 vs 61.6).
- Claude Opus 4.5 (thinking) is stronger in agents & tools, coding, maths.
- GPT-5.2 is stronger in composite, instruction following, knowledge, long context, multimodal, reasoning, human preference.
- GPT-5.2 is 2.1× cheaper ($4.81 vs $10.00 per 1M blended).
- GPT-5.2 streams 1.5× faster (68 vs 45 tokens per second).
| Metric | Claude Opus 4.5 (thinking) | GPT-5.2 |
|---|---|---|
| BenchLeader Index | 60.7 | 61.6 |
| Agents & tools score | 74.6 | 61.2 |
| Coding score | 56.4 | 53.7 |
| Composite score | 65.8 | 67.5 |
| Instruction following score | 54.8 | 67.4 |
| Knowledge score | 61.0 | 61.1 |
| Long context score | 65.2 | 67.9 |
| Maths score | 64.0 | – |
| Multimodal score | 58.0 | 59.9 |
| Reasoning score | 58.6 | 61.3 |
| Human preference score | – | 67.9 |
| Blended price $/M | $10.00 | $4.81 |
| Output speed | 45 tok/s | 68 tok/s |
| Time to first answer | 17.3 s | 129.4 s |
| Context window | 200k | 400k |
| Humanity's Last Exam | – | 27.8% |
| Terminal-Bench | – | 64.9% |
| SimpleBench | – | 45.8% |
| Remote Labor Index | – | 2.1% |
| APEX-Agents | – | 23.0% |
| Epoch Capabilities Index | – | 153.5 |
| LMArena Text | – | 1476 |
| LMArena Hard Prompts | – | 1497 |
| LMArena Coding | – | 1515 |
| LMArena WebDev | – | 1417 |
| LMArena Vision | – | 1268 |
| AA Intelligence Index | 29.1 | 30.4 |
| IFBench | 58.0% | 75.4% |
| AA-LCR | 77.3% | 82.7% |
| MMMU-Pro | 74.0% | – |
| AA-Omniscience | 14 | -0.9 |
| Terminal-Bench Hard | 47.0% | 47.0% |
| GPQA Diamond (AA) | 86.6% | 90.3% |
| Humanity's Last Exam (AA) | 30.1% | 37.7% |
| τ²-Bench Telecom (AA) | 89.5% | 84.8% |
| AIME (Vals) | 95.4% | – |
| LiveCodeBench | 83.7% | – |
| MMLU-Pro | 87.3% | – |
| IOI | 20.3% | – |
| LegalBench | 84.6% | – |
| CorpFin | 65.1% | – |
| TaxEval | 74.9% | – |
| MedQA | 95.9% | – |
| MGSM | 95.2% | – |
| SWE-bench (Vals) | 76.4% | – |
| GPQA Diamond (Vals) | 85.9% | – |
| SWE-Bench Pro | – | 29.9% |
| MultiChallenge | 59.0% | – |
| PRBench Finance | 46.2% | – |
| PRBench Legal | 44.2% | – |
| VISTA | 46.4% | 46.6% |
| MultiNRC | 48.6% | 42.2% |
| TutorBench | 51.2% | 53.5% |
| Kagi LLM Benchmark | 80.2% | 73.3% |
| SWE-bench Verified (bash only) | – | 69.0% |
| ARC-AGI-1 | 80.0% | 94.5% |
| ARC-AGI-2 | 37.6% | 72.9% |
| BFCL Overall | – | 55.9% |
Data as of 2026-09-10. Best configuration of each model; every score links to its source on the model pages.
Claude Opus 4.5 vs GPT-5.2: questions
- Is Claude Opus 4.5 better than GPT-5.2?
- Claude Opus 4.5 (thinking) and GPT-5.2 are level on quality (60.7 vs 61.6). The BenchLeader Index combines every independent quality benchmark; GPT-5.2 is ahead overall as of 2026-09-10, but check the category scores for your use.
- Is Claude Opus 4.5 better than GPT-5.2 for coding?
- Claude Opus 4.5 scores higher in coding (56 vs 54 on the category index, where 50 is average).
- Is Claude Opus 4.5 better than GPT-5.2 for agentic tasks?
- Claude Opus 4.5 scores higher in agentic tasks (75 vs 61 on the category index, where 50 is average).
- Which is cheaper, Claude Opus 4.5 or GPT-5.2?
- GPT-5.2 is cheaper: $4.81 against $10.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Claude Opus 4.5 or GPT-5.2?
- GPT-5.2 streams faster: 68 against 45 output tokens per second.
- Which has the larger context window?
- GPT-5.2 accepts more context: 400k against 200k tokens.