Model comparison
Claude 3 Opus vs Claude Opus 4.8
Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 29.5 on the Noometry Index.
Last verified . 29 shared benchmarks.
Summary
- They share 29 benchmarks with published results for both. Claude 3 Opus scores higher in 0 categories and Claude Opus 4.8 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 4.8 leads 78.4 to 14.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.7% for Claude 3 Opus and 98.3% for Claude Opus 4.8.
Side by side
| Claude 3 Opus | Claude Opus 4.8 | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 29.5 | 60.7 |
| Released | 2024-02-29 | 2026-05-28 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1M |
| Max output | — | 128K |
| Input $ / M tokens | — | $5 |
| Output $ / M tokens | — | $25 |
| Results tracked | 46 | 65 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4.8 leads
Claude 3 Opus: 32.9 (#267), Claude Opus 4.8: 59.9 (#12)
| Benchmark | Claude 3 Opus | Claude Opus 4.8 |
|---|---|---|
| WeirdML | 19.2% | 82.9% |
| LMArena Coding | 1264 | 1490 |
| DeepSWE | — | 59% |
| FrontierCode | — | 46.5% |
| LMArena WebDev | — | 1556 |
| SciCode | — | 53.5% |
| GSO | — | 47.1% |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 38.6% | — |
| BigCodeBench Complete | 57.4% | — |
| ALE-Bench | — | 1,564 |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Claude Opus 4.8 leads
Claude 3 Opus: 24.6 (#116), Claude Opus 4.8: 47.6 (#11)
| Benchmark | Claude 3 Opus | Claude Opus 4.8 |
|---|---|---|
| APEX-Agents | — | 48.9% |
| OSWorld 2.0 | — | 20.6% |
| Remote Labor Index | — | 8.3% |
| τ²-bench Banking | — | 39.7% |
| Cybench | 10% | — |
| DeepResearch Bench | — | 50.2% |
| PostTrainBench | — | 33.8% |
| GBAEval | — | 70.9% |
| GDP.pdf | — | 24% |
| LMArena Search | — | 1204 |
| METR Time Horizons | 29.5% | — |
| Vending-Bench 2 | — | 5,787 |
Reasoning Claude Opus 4.8 leads
Claude 3 Opus: 14.6 (#324), Claude Opus 4.8: 64.7 (#16)
| Benchmark | Claude 3 Opus | Claude Opus 4.8 |
|---|---|---|
| SimpleBench | 23.5% | 64.8% |
| Chess Puzzles | 5% | 34% |
| EnigmaEval | 0.8% | 23.5% |
| LMArena Hard Prompts | 1245 | 1482 |
| DTBench | 61.6% | 94.9% |
| LMCA | 17% | 57.5% |
| Epoch Capabilities Index | 126.91 | 158.21 |
| ForecastBench | 58.4 | 59.9 |
| ARC-AGI-2 | — | 72.1% |
| Kagi LLM Benchmark | — | 88.8% |
| NYT Connections (extended) | — | 91.1% |
| ARC-AGI-1 | — | 92.5% |
| CritPt | — | 20.9% |
| EBR-Bench | — | 28.6% |
| LiveBench Reasoning | 40.6% | — |
| Mystery Game Puzzles | — | 36% |
| LiveBench Data Analysis | 57.9% | — |
| Surface Evolver Bench | — | 87.5% |
| Bench to the Future 3 | — | 0.14 |
| LiveBench | 49.2% | — |
| WinoGrande | 88.5% | — |
Math Claude Opus 4.8 leads
Claude 3 Opus: 14.8 (#299), Claude Opus 4.8: 78.4 (#13)
| Benchmark | Claude 3 Opus | Claude Opus 4.8 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 98.3% |
| LMArena Math | 1273 | 1487 |
| FrontierMath (Tiers 1-3) | — | 80% |
| FrontierMath Tier 4 | — | 56.1% |
| MathArena Final-Answer Competitions | — | 91.8% |
| ProofBench | — | 69% |
| LiveBench Math | 43.6% | — |
| MATH Level 5 | 37.5% | — |
| FrontierMath (Feb 2025 set) | — | 47.2% |
| FrontierMath Tier 4 (v1) | — | 31.3% |
Knowledge Claude Opus 4.8 leads
Claude 3 Opus: 24.5 (#267), Claude Opus 4.8: 61.3 (#29)
| Benchmark | Claude 3 Opus | Claude Opus 4.8 |
|---|---|---|
| GPQA Diamond | 47.2% | 91% |
| SimpleQA Verified | 12.6% | 53% |
| LMArena Expert | 1223 | 1502 |
| Confabulations | 22.7% | — |
| MMLU | 84.6% | — |
Multimodal Claude Opus 4.8 leads
Claude 3 Opus: 27.1 (#116), Claude Opus 4.8: 42.9 (#26)
| Benchmark | Claude 3 Opus | Claude Opus 4.8 |
|---|---|---|
| LMArena Vision | 1023 | 1294 |
| Blueprint-Bench 2 | — | 14.5% |
| Furniture Assembly | — | 42.5% |
| LMArena Document | — | 1475 |
Multilingual Claude Opus 4.8 leads
Claude 3 Opus: 41.4 (#207), Claude Opus 4.8: 55.2 (#33)
| Benchmark | Claude 3 Opus | Claude Opus 4.8 |
|---|---|---|
| LMArena Non-English | 1258 | 1450 |
| LMArena Chinese | 1248 | 1507 |
| LMArena French | 1275 | 1481 |
| LMArena German | 1258 | 1472 |
| LMArena Japanese | 1204 | 1440 |
| LMArena Korean | 1187 | 1432 |
| LMArena Russian | 1280 | 1474 |
| LMArena Spanish | 1246 | 1466 |
Instruction Following Claude Opus 4.8 leads
Claude 3 Opus: 64.1 (#228), Claude Opus 4.8: 77.4 (#24)
| Benchmark | Claude 3 Opus | Claude Opus 4.8 |
|---|---|---|
| LMArena Instruction Following | 1248 | 1476 |
| LiveBench Instruction Following | 63.9% | — |
Long Context Claude Opus 4.8 leads
Claude 3 Opus: 38.2 (#202), Claude Opus 4.8: 45.4 (#35)
| Benchmark | Claude 3 Opus | Claude Opus 4.8 |
|---|---|---|
| LMArena Longer Query | 1259 | 1483 |
Writing & Preference Claude Opus 4.8 leads
Claude 3 Opus: 47.2 (#213), Claude Opus 4.8: 72.0 (#16)
| Benchmark | Claude 3 Opus | Claude Opus 4.8 |
|---|---|---|
| LMArena Text | 1262 | 1461 |
| LMArena Creative Writing | 1235 | 1454 |
| LMArena Multi-Turn | 1275 | 1476 |
| EQ-Bench Creative Writing | — | 1840 |
| EQ-Bench 4 | — | 1281 |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than Claude Opus 4.8?
Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 29.5 on the Noometry Index.
Is Claude 3 Opus or Claude Opus 4.8 better for coding?
Claude Opus 4.8 scores higher on coding benchmarks: 59.9 versus 32.9 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and Claude Opus 4.8 share?
29 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and Claude Opus 4.8 has 65.