Claude Opus 5
| Released | 2026-07-24 |
|---|---|
| Price per 1M | $5.00 in · $25.00 out |
| Overall score | 63.9 · rank 8 |
| Context window | 1,000,000 tokens |
| Provider | Anthropic |
| Weights | Proprietary |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 63.9 | 8 | 53 |
| Overall: Coding | 63.0 | 10 | 38 |
| Overall: Reasoning | 64.8 | 9 | 18 |
| Overall: Agentic work | 63.1 | 10 | 36 |
| Overall: Frontend | 62.4 | 10 | 16 |
| Coding | 62.9 | 11 | 12 |
| Agents | 62.5 | 11 | 13 |
| Reasoning | 65.9 | 8 | 11 |
| Math | 64.2 | 14 | 4 |
| Knowledge | 63.5 | 11 | 3 |
| Frontend | 62.1 | 9 | 2 |
| General | 67.9 | 2 | 6 |
| Multimodal | 62.8 | 4 | 2 |
| Value vs price | +9.2 | 34 |
Buzz
| Posts on X | About 117 a day |
|---|---|
| Mood | 16% positive, 32% negative |
| Hacker News | 52 mentions this week |
| Checked | 2026-11-11 |
- 🇨🇳 DeepSeek and other leading AI companies in China together generate only around one-tenth of the combined revenue of US rivals OpenAI and Anthropic, estimates show, as competition pushes already low Chinese fees further down. Chinese AI companies have gained ground with open-…
- Oh no. Fable 5.2 repeatedly hallucinates contrarian objections that misread you if you make no mistakes. Noooo. This is one of the biggest wastes of time I remember strongly from working with Opus 5 when it came out. Anthropic needs to fix that. Has anyone else seen this? I’ve …
- They built an AI agent in 20 minutes and sold it to Anthropic for $100k - and at Stanford showed how to repeat it from scratch: 02:00 - how to build your first agent with Opus 5 in 20 minutes 22:34 - one agent replaces an Anthropic engineering team worth $200k a year 49:47 - …
Configurations
| Configuration | Balanced | Benchmarks |
|---|---|---|
| Unspecified setting | 64.0 ±2.1 | 11 |
| High | 62.7 ±1.5 | 17 |
| Low | 58.7 ±2.5 | 9 |
| Max | 63.3 ±0.7 | 37 |
| Medium | 61.6 ±3.1 | 6 |
| Extra High | 63.7 ±2.2 | 9 |
Evidence
Coding 62.9 ±1.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| CursorBench | 46.6% | Epoch AI | 2026-11-11 |
| FrontierSWE | 52.0% | Epoch AI | 2026-11-11 |
| DeepSWE | 73.7% | Epoch AI | 2026-11-11 |
| FrontierCode | 53.4% | Epoch AI | 2026-11-11 |
| SWE-rebench | 63.4% | SWE-rebench | 2026-07-01 |
| WeirdML v3 | 19.0% | Epoch AI | 2026-11-11 |
| MirrorCode | 87.8% | Epoch AI | 2026-10-20 |
| IOI | 84.3% | Vals AI | 2026-10-29 |
| SWE-bench Verified (Vals run) | 97.0% | Vals AI | 2026-09-01 |
| SciCode | 56.4% | Epoch AI | 2026-11-11 |
| LiveCodeBench (Vals run) | 89.0% | Vals AI | 2026-09-01 |
| ProgramBench | 3.0% | Vals AI | 2026-11-09 |
Agents 62.5 ±0.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Terminal-Bench 4 | 53.3% | Terminal-Bench | 2026-07-24 |
| OSWorld 2.0 | 31.4% | Epoch AI | 2026-11-11 |
| Terminal-Bench 4.0 (Vals run) | 53.5% | Vals AI | 2026-11-09 |
| SRE Bench | 12.2% | Vals AI | 2026-11-05 |
| Terminal-Bench-Science | 27.1% | Vals AI | 2026-11-09 |
| APEX-Agents | 65.8% | Epoch AI | 2026-11-11 |
| Terminal-Bench 2.1 (Vals run) | 84.6% | Vals AI | 2026-10-01 |
| OSWorld-Verified | 83.4% | OSWorld-Verified | 2026-08-01 |
| tau2-bench Banking Knowledge | 48.7% | tau2-bench | 2026-08-03 |
| Tax Agent Bench | 47.1% | Vals AI | 2026-11-09 |
| SkillsBench | 60.4% | Vals AI | 2026-10-29 |
| BALROG | 63.4% | Epoch AI | 2026-11-11 |
| Vending-Bench 2 | $11,182 | Epoch AI | 2026-11-11 |
| CyberBench v1.1 | 65.4% | Vals AI | 2026-11-05 |
Reasoning 65.9 ±1.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| ARC-AGI-2 | 90.4% | Epoch AI | 2026-11-11 |
| ARC-AGI-2 (ARC Prize verified) | 90.4% | ARC Prize | 2026-11-11 |
| Extended NYT Connections | 94.3% | Lech Mazur | 2026-11-10 |
| MysteryMechanism | 37.4% | Vals AI | 2026-11-09 |
| CritPt | 29.1% | Epoch AI | 2026-11-11 |
| ARC-AGI-1 (ARC Prize verified) | 97.5% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 | 97.5% | Epoch AI | 2026-11-11 |
| GPQA Diamond | 93.9% | Epoch AI | 2026-07-24 |
| GPQA Diamond (Vals run) | 93.4% | Vals AI | 2026-09-01 |
| BioMysteryBench | 79.3% | Vals AI | 2026-11-09 |
| ARC-AGI-3 | 30.2% | ARC Prize | 2026-11-11 |
| Chess Puzzles | 43.0% | Epoch AI | 2026-10-13 |
| SimpleBench | 80.6% | SimpleBench | 2026-07-24 |
| SimpleBench | 80.6% | Epoch AI | 2026-11-11 |
| Mystery Game Puzzles | 59.0% | Epoch AI | 2026-07-25 |
Math 64.2 ±2.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| FrontierMath Tiers 1-3 | 85.6% | Epoch AI | 2026-07-24 |
| FrontierMath Tier 4 | 73.2% | Epoch AI | 2026-07-24 |
| OTIS Mock AIME 2024-2025 | 98.9% | Epoch AI | 2026-07-24 |
| ProofBench v1.1 | 99.0% | Vals AI | 2026-11-09 |
| ProofBench | 99.0% | Epoch AI | 2026-11-11 |
Knowledge 63.5 ±3.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| SimpleQA Verified | 59.9% | Epoch AI | 2026-08-10 |
| MMLU-Pro (Vals run) | 91.6% | Vals AI | 2026-09-01 |
| Legal Research Bench | 55.3% | Vals AI | 2026-11-09 |
Frontend 62.1 ±0.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Vibe Code Bench v1.1 | 88.4% | Vals AI | 2026-11-09 |
| LMArena WebDev | 1,691 | LMArena | 2026-11-09 |
General 67.9 ±2.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LLM Debate Benchmark | 1,717 | Lech Mazur | 2026-10-26 |
| EQ-Bench 4 | 1,365 | EQ-Bench | 2026-10-24 |
| EQ-Bench Creative Writing v3 | 2,135 | EQ-Bench | 2026-11-11 |
| LMArena Text | 1,490 | LMArena | 2026-11-09 |
| Creative Story-Writing (Lech Mazur) | 93.0% | Lech Mazur | 2026-10-23 |
| EQ-Bench Longform Creative Writing | 86.3% | EQ-Bench | 2026-11-11 |
Multimodal 62.8 ±2.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MMMU-Pro (Vals run) | 89.9% | Vals AI | 2026-09-01 |
| LMArena Vision | 1,285 | LMArena | 2026-11-09 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| ARC-AGI-1 | 97.5% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 97.5% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 | 88.3% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 88.3% | ARC Prize | A stronger configuration result was used |
| Chess Puzzles | 27.0% | Epoch AI | A stronger configuration result was used |
| Creative Story-Writing (Lech Mazur) | 90.0% | Lech Mazur | A stronger configuration result was used |
| CritPt | 23.1% | Epoch AI | A stronger configuration result was used |
| CritPt | 27.7% | Epoch AI | A stronger configuration result was used |
| CritPt | 28.3% | Epoch AI | A stronger configuration result was used |
| CritPt | 26.9% | Epoch AI | A stronger configuration result was used |
| CursorBench | 40.7% | Epoch AI | A stronger configuration result was used |
| CursorBench | 46.1% | Epoch AI | A stronger configuration result was used |
| CursorBench | 44.7% | Epoch AI | A stronger configuration result was used |
| CursorBench | 43.3% | Epoch AI | A stronger configuration result was used |
| DeepSWE | 58.1% | Epoch AI | A stronger configuration result was used |
| DeepSWE | 73.2% | Epoch AI | A stronger configuration result was used |
| DeepSWE | 72.8% | Epoch AI | A stronger configuration result was used |
| DeepSWE | 68.9% | Epoch AI | A stronger configuration result was used |
| Extended NYT Connections | 92.2% | Lech Mazur | A stronger configuration result was used |
| GPQA Diamond | 92.9% | Epoch AI | A stronger configuration result was used |
| GPQA Diamond | 87.9% | Epoch AI | A stronger configuration result was used |
| LMArena Text | 1,489 | LMArena | A stronger configuration result was used |
| LMArena WebDev | 1,657 | LMArena | A stronger configuration result was used |
| Mystery Game Puzzles | 37.0% | Epoch AI | A stronger configuration result was used |
| OSWorld 2.0 | 22.3% | Epoch AI | A stronger configuration result was used |
| OSWorld 2.0 | 30.2% | Epoch AI | A stronger configuration result was used |
| OSWorld 2.0 | 29.0% | Epoch AI | A stronger configuration result was used |
| OSWorld 2.0 | 25.3% | Epoch AI | A stronger configuration result was used |
| OTIS Mock AIME 2024-2025 | 97.8% | Epoch AI | A stronger configuration result was used |
| OTIS Mock AIME 2024-2025 | 93.3% | Epoch AI | A stronger configuration result was used |
| SciCode | 48.0% | Epoch AI | A stronger configuration result was used |
| SciCode | 55.0% | Epoch AI | A stronger configuration result was used |
| SciCode | 54.3% | Epoch AI | A stronger configuration result was used |
| SciCode | 50.7% | Epoch AI | A stronger configuration result was used |
| Terminal-Bench 4 | 34.9% | Terminal-Bench | A stronger configuration result was used |
| Terminal-Bench 4 | 52.1% | Terminal-Bench | A stronger configuration result was used |
| Terminal-Bench 4 | 49.7% | Terminal-Bench | A stronger configuration result was used |
| Terminal-Bench 4 | 45.2% | Terminal-Bench | A stronger configuration result was used |
Names matched to this model
claude-opus-5