Claude Opus 4.8
| Released | 2026-05-28 |
|---|---|
| Price per 1M | $5.00 in · $25.00 out |
| Overall score | 60.6 · rank 22 |
| Context window | 1,000,000 tokens |
| Provider | Anthropic |
| Weights | Proprietary |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 60.6 | 22 | 48 |
| Overall: Coding | 60.2 | 24 | 32 |
| Overall: Reasoning | 61.3 | 26 | 19 |
| Overall: Agentic work | 60.3 | 31 | 30 |
| Overall: Frontend | 59.7 | 25 | 12 |
| Coding | 60.2 | 25 | 8 |
| Agents | 59.9 | 33 | 11 |
| Reasoning | 62.0 | 27 | 11 |
| Math | 61.3 | 21 | 5 |
| Knowledge | 59.5 | 26 | 3 |
| Frontend | 59.5 | 26 | 2 |
| General | 62.3 | 17 | 6 |
| Multimodal | 59.6 | 21 | 2 |
| Value vs price | +5.9 | 64 |
Configurations
| Configuration | Balanced | Benchmarks |
|---|---|---|
| Unspecified setting | 59.7 ±2.2 | 13 |
| High | 57.9 ±2.1 | 11 |
| Low | 56.8 ±2.7 | 7 |
| Max | 60.4 ±1.0 | 31 |
| Medium | 56.7 ±3.4 | 4 |
| Non-reasoning | 53.3 ±5.5 | 3 |
| Thinking | 67.0 ±9.4 | 1 |
| Extra High | 59.5 ±3.9 | 5 |
Evidence
Coding 60.2 ±1.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| DeepSWE | 59.0% | Epoch AI | 2026-11-11 |
| FrontierCode | 46.5% | Epoch AI | 2026-11-11 |
| MirrorCode | 65.6% | Epoch AI | 2026-10-21 |
| GSO | 47.1% | Epoch AI | 2026-11-11 |
| SWE-bench Verified (Vals run) | 88.6% | Vals AI | 2026-09-01 |
| SciCode | 53.5% | Epoch AI | 2026-11-11 |
| LiveCodeBench (Vals run) | 87.8% | Vals AI | 2026-09-01 |
| ProgramBench | 1.0% | Vals AI | 2026-11-09 |
Agents 59.9 ±0.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| OSWorld 2.0 | 20.6% | Epoch AI | 2026-11-11 |
| Terminal-Bench 4 | 23.6% | Terminal-Bench | 2026-05-28 |
| Terminal-Bench 4.0 (Vals run) | 23.2% | Vals AI | 2026-11-09 |
| Terminal-Bench-Science | 4.3% | Vals AI | 2026-11-09 |
| APEX-Agents | 48.9% | Epoch AI | 2026-11-11 |
| Remote Labor Index | 8.3% | Epoch AI | 2026-11-11 |
| Terminal-Bench 2.1 (Vals run) | 71.9% | Vals AI | 2026-10-01 |
| tau2-bench Banking Knowledge | 39.7% | tau2-bench | 2026-07-23 |
| Tax Agent Bench | 30.0% | Vals AI | 2026-11-09 |
| SkillsBench | 59.2% | Vals AI | 2026-10-29 |
| DeepResearch Bench | 50.2% | Epoch AI | 2026-11-11 |
| Vending-Bench 2 | $5,787 | Epoch AI | 2026-11-11 |
Reasoning 62.0 ±1.4
| Benchmark | Result | Source | Date |
|---|---|---|---|
| ARC-AGI-2 | 72.1% | Epoch AI | 2026-11-11 |
| ARC-AGI-2 (ARC Prize verified) | 72.1% | ARC Prize | 2026-11-11 |
| Extended NYT Connections | 91.1% | Lech Mazur | 2026-11-10 |
| CritPt | 20.9% | Epoch AI | 2026-11-11 |
| ARC-AGI-1 (ARC Prize verified) | 92.5% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 | 92.5% | Epoch AI | 2026-11-11 |
| EnigmaEval | 23.5% | Epoch AI | 2026-11-11 |
| GPQA Diamond | 91.0% | Epoch AI | 2026-06-07 |
| GPQA Diamond (Vals run) | 92.4% | Vals AI | 2026-09-01 |
| ARC-AGI-3 | 1.5% | ARC Prize | 2026-11-11 |
| Kagi Reasoning Benchmark | 88.8% | Kagi | 2026-11-11 |
| Chess Puzzles | 52.0% | Epoch AI | 2026-10-13 |
| SimpleBench | 64.8% | SimpleBench | 2026-05-29 |
| SimpleBench | 64.8% | Epoch AI | 2026-11-11 |
| Mystery Game Puzzles | 36.0% | Epoch AI | 2026-07-25 |
Math 61.3 ±1.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| FrontierMath Tiers 1-3 | 80.0% | Epoch AI | 2026-06-10 |
| FrontierMath Tier 4 | 56.1% | Epoch AI | 2026-06-10 |
| ProofBench | 69.0% | Epoch AI | 2026-11-11 |
| OTIS Mock AIME 2024-2025 | 98.3% | Epoch AI | 2026-06-07 |
| MathArena Final-Answer Competitions | 91.8% | MathArena | 2026-11-11 |
| FrontierMath Tiers 1-3 (2025 set) | 47.2% | Epoch AI | 2026-06-08 |
| FrontierMath Tier 4 (2025 set) | 31.3% | Epoch AI | 2026-06-08 |
Knowledge 59.5 ±3.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| SimpleQA Verified | 53.0% | Epoch AI | 2026-08-27 |
| MMLU-Pro (Vals run) | 89.6% | Vals AI | 2026-09-01 |
| Legal Research Bench | 43.8% | Vals AI | 2026-11-09 |
Frontend 59.5 ±0.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Vibe Code Bench v1.1 | 82.7% | Vals AI | 2026-11-09 |
| LMArena WebDev | 1,555 | LMArena | 2026-11-09 |
General 62.3 ±1.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LLM Debate Benchmark | 1,624 | Lech Mazur | 2026-10-26 |
| EQ-Bench 4 | 1,268 | EQ-Bench | 2026-10-24 |
| EQ-Bench Creative Writing v3 | 1,840 | EQ-Bench | 2026-11-11 |
| LMArena Text | 1,482 | LMArena | 2026-11-09 |
| Creative Story-Writing (Lech Mazur) | 61.0% | Lech Mazur | 2026-10-23 |
| EQ-Bench Longform Creative Writing | 80.8% | EQ-Bench | 2026-11-11 |
Multimodal 59.6 ±2.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MMMU-Pro (Vals run) | 86.6% | Vals AI | 2026-09-01 |
| LMArena Vision | 1,286 | LMArena | 2026-11-09 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| ARC-AGI-1 | 88.0% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 | 92.0% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 | 91.5% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 88.0% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 92.0% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 91.5% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 | 62.2% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 | 71.7% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 62.2% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 71.7% | ARC Prize | A stronger configuration result was used |
| Chess Puzzles | 19.0% | Epoch AI | A stronger configuration result was used |
| Chess Puzzles | 29.0% | Epoch AI | A stronger configuration result was used |
| Creative Story-Writing (Lech Mazur) | 53.0% | Lech Mazur | A stronger configuration result was used |
| DeepResearch Bench | 49.3% | Epoch AI | A stronger configuration result was used |
| DeepResearch Bench | 47.4% | Epoch AI | A stronger configuration result was used |
| DeepSWE | 40.8% | Epoch AI | A stronger configuration result was used |
| DeepSWE | 54.4% | Epoch AI | A stronger configuration result was used |
| DeepSWE | 51.8% | Epoch AI | A stronger configuration result was used |
| DeepSWE | 48.7% | Epoch AI | A stronger configuration result was used |
| Extended NYT Connections | 88.3% | Lech Mazur | A stronger configuration result was used |
| GPQA Diamond | 85.4% | Epoch AI | A stronger configuration result was used |
| GPQA Diamond | 88.4% | Epoch AI | A stronger configuration result was used |
| LMArena Text | 1,474 | LMArena | A stronger configuration result was used |
| LMArena Vision | 1,279 | LMArena | A stronger configuration result was used |
| LMArena WebDev | 1,536 | LMArena | A stronger configuration result was used |
| Mystery Game Puzzles | 31.0% | Epoch AI | A stronger configuration result was used |
| OTIS Mock AIME 2024-2025 | 84.4% | Epoch AI | A stronger configuration result was used |
| OTIS Mock AIME 2024-2025 | 97.8% | Epoch AI | A stronger configuration result was used |
| SWE-bench Verified (Vals run) | 85.8% | Vals AI | A stronger configuration result was used |
| Terminal-Bench 2.1 (Vals run) | 69.7% | Vals AI | A stronger configuration result was used |
| Vending-Bench 2 | $2,992 | Epoch AI | A stronger configuration result was used |
| Vibe Code Bench v1.1 | 77.5% | Vals AI | A stronger configuration result was used |
Names matched to this model
claude-opus-4-8