Model comparison
Claude 3 Haiku vs Command A
Command A is the stronger model overall, scoring 36.5 to 25.9 on the Noometry Index.
Last verified . 21 shared benchmarks.
CohereSummary
- They share 21 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Command A in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Command A leads 36.2 to 9.8.
- The biggest single-benchmark swing is DTBench: 50.1% for Claude 3 Haiku and 61.3% for Command A.
- Command A has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Haiku | Command A | |
|---|---|---|
| Provider | Anthropic | Cohere |
| Noometry Index | 25.9 | 36.5 |
| Released | 2024-03-07 | 2025-03-13 |
| Weights | Proprietary | Open |
| Context window | — | 256K |
| Max output | — | 8K |
| Input $ / M tokens | — | $2.50 |
| Output $ / M tokens | — | $10 |
| Results tracked | 37 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Claude 3 Haiku: 26.4 (#325), Command A: 27.2 (#322)
| Benchmark | Claude 3 Haiku | Command A |
|---|---|---|
| LMArena Coding | 1199 | 1330 |
| Aider Polyglot | — | 12% |
| WeirdML | 9.8% | — |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |
Agentic & Tool Use Not comparable
Claude 3 Haiku: —, Command A: 35.9 (#40)
| Benchmark | Claude 3 Haiku | Command A |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 57.1% |
Reasoning Command A leads
Claude 3 Haiku: 16.3 (#307), Command A: 18.3 (#283)
| Benchmark | Claude 3 Haiku | Command A |
|---|---|---|
| Kagi LLM Benchmark | 34.2% | 28.8% |
| LMArena Hard Prompts | 1174 | 1326 |
| DTBench | 50.1% | 61.3% |
| LMCA | 8.8% | 10.3% |
| Epoch Capabilities Index | 118.35 | — |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |
Math Command A leads
Claude 3 Haiku: 9.8 (#319), Command A: 36.2 (#171)
| Benchmark | Claude 3 Haiku | Command A |
|---|---|---|
| LMArena Math | 1188 | 1300 |
| OTIS Mock AIME 2024-2025 | 1.8% | — |
| MATH Level 5 | 14.9% | — |
Knowledge Command A leads
Claude 3 Haiku: 17.3 (#285), Command A: 37.1 (#159)
| Benchmark | Claude 3 Haiku | Command A |
|---|---|---|
| LMArena Expert | 1148 | 1295 |
| GPQA Diamond | 36.3% | — |
| Confabulations | 34.2% | — |
| Vectara Hallucination Rate | — | 9.3% |
| MMLU | 73.8% | — |
Multimodal Not comparable
Claude 3 Haiku: 23.6 (#128), Command A: —
| Benchmark | Claude 3 Haiku | Command A |
|---|---|---|
| LMArena Vision | 950 | — |
| ScienceQA | 72% | — |
Multilingual Command A leads
Claude 3 Haiku: 36.0 (#243), Command A: 45.3 (#170)
| Benchmark | Claude 3 Haiku | Command A |
|---|---|---|
| LMArena Non-English | 1178 | 1313 |
| LMArena Chinese | 1155 | 1327 |
| LMArena French | 1195 | 1351 |
| LMArena German | 1174 | 1341 |
| LMArena Japanese | 1102 | 1285 |
| LMArena Korean | 1109 | 1285 |
| LMArena Russian | 1204 | 1314 |
| LMArena Spanish | 1166 | 1347 |
Instruction Following Command A leads
Claude 3 Haiku: 61.3 (#247), Command A: 69.1 (#177)
| Benchmark | Claude 3 Haiku | Command A |
|---|---|---|
| LMArena Instruction Following | 1173 | 1309 |
Long Context Command A leads
Claude 3 Haiku: 36.1 (#237), Command A: 40.6 (#151)
| Benchmark | Claude 3 Haiku | Command A |
|---|---|---|
| LMArena Longer Query | 1190 | 1334 |
Writing & Preference Command A leads
Claude 3 Haiku: 29.7 (#291), Command A: 47.6 (#208)
| Benchmark | Claude 3 Haiku | Command A |
|---|---|---|
| LMArena Text | 1195 | 1331 |
| LMArena Creative Writing | 1157 | 1319 |
| EQ-Bench Creative Writing | 717 | 1145 |
| LMArena Multi-Turn | 1190 | 1339 |
Frequently asked questions
Is Claude 3 Haiku better than Command A?
Command A is the stronger model overall, scoring 36.5 to 25.9 on the Noometry Index.
Is Claude 3 Haiku or Command A better for coding?
They score almost the same on coding (26.4 vs 27.2); test both on your own repository before choosing.
How many benchmarks do Claude 3 Haiku and Command A share?
21 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Command A has 24.