Model comparison
Claude 3 Haiku vs Claude Opus 4.1
Claude Opus 4.1 is the stronger model overall, scoring 40.8 to 25.7 on the Noometry Index.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Claude Opus 4.1 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Claude Opus 4.1 leads 62.6 to 29.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 68.9% for Claude Opus 4.1.
Side by side
| Claude 3 Haiku | Claude Opus 4.1 | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 25.7 | 40.8 |
| Released | 2024-03-07 | 2025-08-05 |
| Weights | Proprietary | Proprietary |
| Context window | — | 200K |
| Max output | — | 32K |
| Input $ / M tokens | — | $15 |
| Output $ / M tokens | — | $75 |
| Results tracked | 37 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4.1 leads
Claude 3 Haiku: 26.1 (#320), Claude Opus 4.1: 44.3 (#69)
| Benchmark | Claude 3 Haiku | Claude Opus 4.1 |
|---|---|---|
| WeirdML | 9.8% | 45.9% |
| LMArena Coding | 1199 | 1479 |
| SWE-bench Verified | — | 73.3% |
| LMArena WebDev | — | 1390 |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| ALE-Bench | — | 674.77 |
| AlgoTune | — | 1.34 |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |
Agentic & Tool Use Not comparable
Claude 3 Haiku: —, Claude Opus 4.1: 34.8 (#38)
| Benchmark | Claude 3 Haiku | Claude Opus 4.1 |
|---|---|---|
| Terminal-Bench | — | 38% |
| GDPval | — | 43.6% |
| Cybench | — | 42% |
| DeepResearch Bench | — | 48.3% |
| LMArena Search | — | 1148 |
| METR Time Horizons | — | 66.8% |
Reasoning Claude Opus 4.1 leads
Claude 3 Haiku: 15.7 (#303), Claude Opus 4.1: 31.6 (#70)
| Benchmark | Claude 3 Haiku | Claude Opus 4.1 |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1443 |
| DTBench | 50.1% | 80% |
| LMCA | 8.8% | 37.1% |
| Epoch Capabilities Index | 118.3 | 144.11 |
| ForecastBench | 53.2 | 62 |
| SimpleBench | — | 60% |
| Kagi LLM Benchmark | 34.2% | — |
| Chess Puzzles | — | 7% |
| EnigmaEval | — | 7.2% |
| EBR-Bench | — | 7.9% |
| Mystery Game Puzzles | — | 21% |
| WinoGrande | 74.2% | — |
Math Claude Opus 4.1 leads
Claude 3 Haiku: 9.2 (#313), Claude Opus 4.1: 22.0 (#267)
| Benchmark | Claude 3 Haiku | Claude Opus 4.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 68.9% |
| LMArena Math | 1188 | 1430 |
| FrontierMath (Tiers 1-3) | — | 12.6% |
| FrontierMath Tier 4 | — | 2.4% |
| MATH Level 5 | 14.9% | — |
| FrontierMath (Feb 2025 set) | — | 7.2% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Claude Opus 4.1 leads
Claude 3 Haiku: 17.3 (#280), Claude Opus 4.1: 41.9 (#93)
| Benchmark | Claude 3 Haiku | Claude Opus 4.1 |
|---|---|---|
| GPQA Diamond | 36.3% | 77.3% |
| Confabulations | 34.2% | 17.1% |
| LMArena Expert | 1148 | 1439 |
| Humanity's Last Exam | — | 11.5% |
| Vectara Hallucination Rate | — | 11.8% |
| MMLU | 73.8% | — |
Multimodal Claude Opus 4.1 leads
Claude 3 Haiku: 23.1 (#123), Claude Opus 4.1: 26.2 (#114)
| Benchmark | Claude 3 Haiku | Claude Opus 4.1 |
|---|---|---|
| LMArena Vision | 950 | — |
| VPCT | — | 35% |
| ScienceQA | 72% | — |
Multilingual Claude Opus 4.1 leads
Claude 3 Haiku: 36.1 (#238), Claude Opus 4.1: 52.1 (#90)
| Benchmark | Claude 3 Haiku | Claude Opus 4.1 |
|---|---|---|
| LMArena Non-English | 1178 | 1405 |
| LMArena Chinese | 1155 | 1427 |
| LMArena French | 1194 | 1432 |
| LMArena German | 1174 | 1412 |
| LMArena Japanese | 1102 | 1378 |
| LMArena Korean | 1108 | 1381 |
| LMArena Russian | 1204 | 1421 |
| LMArena Spanish | 1165 | 1448 |
Instruction Following Claude Opus 4.1 leads
Claude 3 Haiku: 61.5 (#242), Claude Opus 4.1: 75.8 (#53)
| Benchmark | Claude 3 Haiku | Claude Opus 4.1 |
|---|---|---|
| LMArena Instruction Following | 1174 | 1436 |
Long Context Claude Opus 4.1 leads
Claude 3 Haiku: 36.1 (#232), Claude Opus 4.1: 44.5 (#60)
| Benchmark | Claude 3 Haiku | Claude Opus 4.1 |
|---|---|---|
| LMArena Longer Query | 1191 | 1455 |
Writing & Preference Claude Opus 4.1 leads
Claude 3 Haiku: 29.8 (#286), Claude Opus 4.1: 62.6 (#70)
| Benchmark | Claude 3 Haiku | Claude Opus 4.1 |
|---|---|---|
| LMArena Text | 1195 | 1419 |
| LMArena Creative Writing | 1157 | 1412 |
| LMArena Multi-Turn | 1190 | 1444 |
| Short-Story Creative Writing | — | 84.7% |
| EQ-Bench Creative Writing | 717 | — |
Frequently asked questions
Is Claude 3 Haiku better than Claude Opus 4.1?
Claude Opus 4.1 is the stronger model overall, scoring 40.8 to 25.7 on the Noometry Index.
Is Claude 3 Haiku or Claude Opus 4.1 better for coding?
Claude Opus 4.1 scores higher on coding benchmarks: 44.3 versus 26.1 in the Noometry coding category.
How many benchmarks do Claude 3 Haiku and Claude Opus 4.1 share?
25 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Claude Opus 4.1 has 48.