Claude Opus 4.1 vs Claude Opus 5
Claude Opus 4.1 Verdict
- Claude Opus 5 leads on quality: 70.0 vs 58.2.
- Claude Opus 4.1 (thinking) is stronger in instruction following.
- Claude Opus 5 is stronger in agents & tools, coding, composite, human preference, knowledge, long context, maths, multimodal, reasoning.
- Claude Opus 5 is 3.0× cheaper ($10.00 vs $30.00 per 1M blended).
- Claude Opus 5 streams 5.1× faster (51 vs 10 tokens per second).
| Metric | Claude Opus 4.1 (thinking) | Claude Opus 5 |
|---|---|---|
| BenchLeader Index | 58.2 | 70.0 |
| Agents & tools score | 63.6 | 71.5 |
| Coding score | 53.3 | 74.5 |
| Composite score | 57.9 | 93.1 |
| Human preference score | 64.6 | 78.2 |
| Instruction following score | 52.6 | – |
| Knowledge score | 58.3 | 70.6 |
| Long context score | 64.5 | 66.2 |
| Maths score | 59.3 | 67.1 |
| Multimodal score | 56.5 | 69.4 |
| Reasoning score | 65.6 | 71.5 |
| Blended price $/M | $30.00 | $10.00 |
| Output speed | 10 tok/s | 51 tok/s |
| Time to first answer | 3.2 s | 93.0 s |
| Context window | 200k | 1M |
| GPQA Diamond | – | 92.9% |
| OTIS Mock AIME | – | 97.8% |
| SimpleBench | – | 80.6% |
| WeirdML | – | 86.3% |
| Epoch Capabilities Index | – | 162.6 |
| LMArena Text | 1450 | – |
| LMArena Hard Prompts | 1480 | – |
| LMArena Coding | 1512 | – |
| AA Intelligence Index | 22.9 | 50.7 |
| IFBench | 55.4% | – |
| AA-LCR | 76.0% | 79.3% |
| MMMU-Pro | 67.9% | 84.7% |
| AA-Omniscience | – | 37.1 |
| Terminal-Bench Hard | 34.3% | – |
| GPQA Diamond (AA) | 80.9% | 93.2% |
| Humanity's Last Exam (AA) | 12.5% | 54.9% |
| SciCode (AA) | – | 56.4% |
| τ²-Bench Telecom (AA) | 71.4% | – |
| AIME (Vals) | 78.2% | – |
| LiveCodeBench | 66.5% | 89.0% |
| MMLU-Pro | 87.9% | 91.6% |
| IOI | – | 91.7% |
| LegalBench | – | 87.0% |
| CorpFin | – | 73.2% |
| TaxEval | 73.7% | 75.1% |
| MedQA | 93.6% | – |
| MGSM | 94.4% | – |
| Terminal-Bench 2.1 (Vals) | – | 84.6% |
| SWE-bench (Vals) | – | 97.0% |
| GPQA Diamond (Vals) | 76.3% | 93.4% |
| Vals Index | – | 67.2 |
| MultiChallenge | 57.2% | – |
| VISTA | 48.4% | – |
| MultiNRC | 38.4% | – |
| HiL-Bench | – | 57.0% |
| TutorBench | 50.8% | – |
| EQ-Bench 4 | – | 1385 |
Data as of 2026-09-10. Best configuration of each model; every score links to its source on the model pages.
Claude Opus 4.1 vs Claude Opus 5: questions
- Is Claude Opus 4.1 better than Claude Opus 5?
- Claude Opus 5 leads on quality: 70.0 vs 58.2. The BenchLeader Index combines every independent quality benchmark; Claude Opus 5 is ahead overall as of 2026-09-10, but check the category scores for your use.
- Is Claude Opus 4.1 better than Claude Opus 5 for coding?
- Claude Opus 5 scores higher in coding (75 vs 53 on the category index, where 50 is average).
- Is Claude Opus 4.1 better than Claude Opus 5 for agentic tasks?
- Claude Opus 5 scores higher in agentic tasks (72 vs 64 on the category index, where 50 is average).
- Which is cheaper, Claude Opus 4.1 or Claude Opus 5?
- Claude Opus 5 is cheaper: $10.00 against $30.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Claude Opus 4.1 or Claude Opus 5?
- Claude Opus 5 streams faster: 51 against 10 output tokens per second.
- Which has the larger context window?
- Claude Opus 5 accepts more context: 1M against 200k tokens.