Claude Opus 4.8 vs Claude Opus 5
Different efforts: max · medium · Compare all at high
Humanity's Last Exam Humanity's Last Exam: Very hard expert questions across many fields, written to stump today’s models
Opus 4.848.7
Opus 551.3
Terminal-Bench 4.0 Terminal-Bench 4.0 · source setting 2: Agents use a terminal to complete executable coding, system administration and research tasks in the 4.0 suite
Opus 4.821.7
Opus 534.3
AutomationBench-AA AutomationBench-AA: Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation
Opus 4.845.6
Opus 554.3
Arena Text · creative writing Arena Text · creative writing: Human pairwise preferences in the stated text vote pool
Opus 4.8 · high1,469
Opus 5 · max1,471
MMMU-Pro (vision) MMMU-Pro (vision): College-level questions that require understanding images: charts, diagrams, photos
Opus 4.8—
Opus 581.6
| Benchmark | Opus 4.8 | Opus 5 |
|---|---|---|
| Humanity's Last Exam | 48.7 | 51.3 |
| Terminal-Bench 4.0 | 21.7 | 34.3 |
| AutomationBench-AA | 45.6 | 54.3 |
| Arena Text · creative writing | 1,469 · high | 1,471 · max |
| MMMU-Pro (vision) | — | 81.6 |
Specs
| Spec | Opus 4.8 | Opus 5 |
|---|---|---|
| Input, per 1M | $5 | $5 |
| Output, per 1M | $25 | $25 |
| Cost per task? | $4.08 | $2.19 |
| Speed, tok/s | 66 | 55 |
| Context | 1M | 1M |
| Weights | Closed | Closed |
| Released | May 2026 | Jul 2026 |
| See images | ✓ | ✓ |
| Read PDFs | Partial | Partial |
| Hear audio | — | — |
| Watch video | — | — |
| Call tools | ✓ | ✓ |
| Force a tool call | Partial | Partial |
| Return strict JSON | ✓ | ✓ |
| Search the web | ? | ? |
| Run code | ? | ? |
| Use a computer | ? | ? |
What breaks when you switch
Opus 4.8 → Opus 5
Nothing known.
Opus 5 → Opus 4.8
Nothing known.