Model comparison
Claude 3 Haiku vs Claude Opus 4.7
Claude Opus 4.7 is the stronger model overall, scoring 58.3 to 25.9 on the Noometry Index.
Last verified . 27 shared benchmarks.
Summary
- They share 27 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Claude Opus 4.7 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 4.7 leads 66.7 to 9.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 97.8% for Claude Opus 4.7.
Side by side
| Claude 3 Haiku | Claude Opus 4.7 | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 25.9 | 58.3 |
| Released | 2024-03-07 | 2026-04-14 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1M |
| Max output | — | 128K |
| Input $ / M tokens | — | $5 |
| Output $ / M tokens | — | $25 |
| Results tracked | 37 | 66 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4.7 leads
Claude 3 Haiku: 26.4 (#325), Claude Opus 4.7: 59.6 (#13)
| Benchmark | Claude 3 Haiku | Claude Opus 4.7 |
|---|---|---|
| WeirdML | 9.8% | 76.4% |
| LMArena Coding | 1199 | 1518 |
| SWE-bench Verified | — | 83.5% |
| FrontierCode | — | 38.5% |
| LMArena WebDev | — | 1558 |
| SciCode | — | 54.5% |
| GSO | — | 44.1% |
| BigCodeBench Instruct | 39.4% | — |
| MirrorCode | — | 31.1% |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| ALE-Bench | — | 1,323 |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |
Agentic & Tool Use Not comparable
Claude 3 Haiku: —, Claude Opus 4.7: 47.9 (#10)
| Benchmark | Claude 3 Haiku | Claude Opus 4.7 |
|---|---|---|
| Terminal-Bench | — | 80.2% |
| APEX-Agents | — | 49.2% |
| OSWorld 2.0 | — | 18.2% |
| τ²-bench Banking | — | 40.2% |
| PostTrainBench | — | 28.6% |
| ExploitBench | — | 26.5% |
| GBAEval | — | 43.8% |
| GDP.pdf | — | 21% |
| LMArena Search | — | 1233 |
| Vending-Bench 2 | — | 10,937 |
Reasoning Claude Opus 4.7 leads
Claude 3 Haiku: 16.3 (#307), Claude Opus 4.7: 53.8 (#29)
| Benchmark | Claude 3 Haiku | Claude Opus 4.7 |
|---|---|---|
| Kagi LLM Benchmark | 34.2% | 80.7% |
| LMArena Hard Prompts | 1174 | 1506 |
| DTBench | 50.1% | 94.7% |
| LMCA | 8.8% | 52.2% |
| Epoch Capabilities Index | 118.35 | 156.25 |
| ForecastBench | 53.2 | 60.3 |
| ARC-AGI-2 | — | 75.8% |
| SimpleBench | — | 61.7% |
| NYT Connections (extended) | — | 39% |
| ARC-AGI-1 | — | 93.5% |
| CritPt | — | 12% |
| Chess Puzzles | — | 30% |
| Thematic Generalization | — | 72.8% |
| EBR-Bench | — | 19% |
| Mystery Game Puzzles | — | 28% |
| WinoGrande | 74.2% | — |
Math Claude Opus 4.7 leads
Claude 3 Haiku: 9.8 (#319), Claude Opus 4.7: 66.7 (#26)
| Benchmark | Claude 3 Haiku | Claude Opus 4.7 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 97.8% |
| LMArena Math | 1188 | 1499 |
| FrontierMath (Tiers 1-3) | — | 70.2% |
| FrontierMath Tier 4 | — | 31.7% |
| MathArena Final-Answer Competitions | — | 73.6% |
| ProofBench | — | 54% |
| MATH Level 5 | 14.9% | — |
| FrontierMath (Feb 2025 set) | — | 43.8% |
| FrontierMath Tier 4 (v1) | — | 22.9% |
Knowledge Claude Opus 4.7 leads
Claude 3 Haiku: 17.3 (#285), Claude Opus 4.7: 62.6 (#23)
| Benchmark | Claude 3 Haiku | Claude Opus 4.7 |
|---|---|---|
| GPQA Diamond | 36.3% | 90.2% |
| LMArena Expert | 1148 | 1521 |
| Humanity's Last Exam | — | 36.2% |
| SimpleQA Verified | — | 51.7% |
| Confabulations | 34.2% | — |
| Vectara Hallucination Rate | — | 12% |
| MMLU | 73.8% | — |
Multimodal Claude Opus 4.7 leads
Claude 3 Haiku: 23.6 (#128), Claude Opus 4.7: 41.2 (#38)
| Benchmark | Claude 3 Haiku | Claude Opus 4.7 |
|---|---|---|
| LMArena Vision | 950 | 1316 |
| Blueprint-Bench 2 | — | 24.5% |
| Furniture Assembly | — | 33.3% |
| LMArena Document | — | 1495 |
| ScienceQA | 72% | — |
Multilingual Claude Opus 4.7 leads
Claude 3 Haiku: 36.0 (#243), Claude Opus 4.7: 57.3 (#10)
| Benchmark | Claude 3 Haiku | Claude Opus 4.7 |
|---|---|---|
| LMArena Non-English | 1178 | 1480 |
| LMArena Chinese | 1155 | 1531 |
| LMArena French | 1195 | 1503 |
| LMArena German | 1174 | 1495 |
| LMArena Japanese | 1102 | 1472 |
| LMArena Korean | 1109 | 1464 |
| LMArena Russian | 1204 | 1494 |
| LMArena Spanish | 1166 | 1495 |
Instruction Following Claude Opus 4.7 leads
Claude 3 Haiku: 61.3 (#247), Claude Opus 4.7: 78.4 (#10)
| Benchmark | Claude 3 Haiku | Claude Opus 4.7 |
|---|---|---|
| LMArena Instruction Following | 1173 | 1498 |
Long Context Claude Opus 4.7 leads
Claude 3 Haiku: 36.1 (#237), Claude Opus 4.7: 46.2 (#25)
| Benchmark | Claude 3 Haiku | Claude Opus 4.7 |
|---|---|---|
| LMArena Longer Query | 1190 | 1505 |
Writing & Preference Claude Opus 4.7 leads
Claude 3 Haiku: 29.7 (#291), Claude Opus 4.7: 75.1 (#8)
| Benchmark | Claude 3 Haiku | Claude Opus 4.7 |
|---|---|---|
| LMArena Text | 1195 | 1490 |
| LMArena Creative Writing | 1157 | 1486 |
| EQ-Bench Creative Writing | 717 | 1914 |
| LMArena Multi-Turn | 1190 | 1505 |
| EQ-Bench 4 | — | 1311 |
Frequently asked questions
Is Claude 3 Haiku better than Claude Opus 4.7?
Claude Opus 4.7 is the stronger model overall, scoring 58.3 to 25.9 on the Noometry Index.
Is Claude 3 Haiku or Claude Opus 4.7 better for coding?
Claude Opus 4.7 scores higher on coding benchmarks: 59.6 versus 26.4 in the Noometry coding category.
How many benchmarks do Claude 3 Haiku and Claude Opus 4.7 share?
27 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Claude Opus 4.7 has 66.