Model comparison
Command A vs Llama 4 Maverick
Command A is the stronger model overall, scoring 36.5 to 30.9 on the Noometry Index. Llama 4 Maverick costs 14× less per token, which makes it the better buy when Command A's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
Cohere
MetaSummary
- They share 24 benchmarks with published results for both. Command A scores higher in 8 categories and Llama 4 Maverick in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Command A leads 36.2 to 26.0.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 28.8% for Command A and 55.9% for Llama 4 Maverick.
- Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $2.50 / $10 for Command A.
- Command A accepts more context: 256K tokens versus 128K.
Side by side
| Command A | Llama 4 Maverick | |
|---|---|---|
| Provider | Cohere | Meta |
| Noometry Index | 36.5 | 30.9 |
| Released | 2025-03-13 | 2025-04-05 |
| Weights | Open | Open |
| Context window | 256K | 128K |
| Max output | 8K | 4K |
| Input $ / M tokens | $2.50 | $0.19 |
| Output $ / M tokens | $10 | $0.65 |
| Results tracked | 24 | 54 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Command A: 27.2 (#322), Llama 4 Maverick: 26.6 (#324)
| Benchmark | Command A | Llama 4 Maverick |
|---|---|---|
| Aider Polyglot | 12% | 15.6% |
| LMArena Coding | 1330 | 1302 |
| SWE-bench Verified (bash only) | — | 21% |
| SciCode | — | 33.1% |
| WeirdML | — | 24.5% |
| BigCodeBench Instruct | — | 49.7% |
| BigCodeBench Complete | — | 61.4% |
| ALE-Bench | — | 172.97 |
Agentic & Tool Use Command A leads
Command A: 35.9 (#40), Llama 4 Maverick: 28.2 (#91)
| Benchmark | Command A | Llama 4 Maverick |
|---|---|---|
| Berkeley Function Calling Leaderboard | 57.1% | 37.3% |
Reasoning Command A leads
Command A: 18.3 (#283), Llama 4 Maverick: 10.1 (#342)
| Benchmark | Command A | Llama 4 Maverick |
|---|---|---|
| Kagi LLM Benchmark | 28.8% | 55.9% |
| LMArena Hard Prompts | 1326 | 1281 |
| DTBench | 61.3% | 61.9% |
| LMCA | 10.3% | 15.9% |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 27.7% |
| NYT Connections (extended) | — | 8% |
| ARC-AGI-1 | — | 4.4% |
| CritPt | — | 0% |
| EnigmaEval | — | 0.6% |
| Epoch Capabilities Index | — | 132.2 |
| ForecastBench | — | 57.5 |
Math Command A leads
Command A: 36.2 (#171), Llama 4 Maverick: 26.0 (#262)
| Benchmark | Command A | Llama 4 Maverick |
|---|---|---|
| LMArena Math | 1300 | 1299 |
| OTIS Mock AIME 2024-2025 | — | 20.6% |
| Omni-MATH | — | 42.2% |
| MATH Level 5 | — | 73% |
| FrontierMath (Feb 2025 set) | — | 0.7% |
Knowledge Command A leads
Command A: 37.1 (#159), Llama 4 Maverick: 33.4 (#204)
| Benchmark | Command A | Llama 4 Maverick |
|---|---|---|
| Vectara Hallucination Rate | 9.3% | 8.2% |
| LMArena Expert | 1295 | 1259 |
| GPQA Diamond | — | 67% |
| Humanity's Last Exam | — | 5.7% |
| MMLU-Pro | — | 81% |
| Confabulations | — | 22.6% |
| GPQA (HELM) | — | 65% |
Multimodal Not comparable
Command A: —, Llama 4 Maverick: 31.6 (#105)
| Benchmark | Command A | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | — | 1142 |
| GeoBench | — | 52% |
| SpatialViz-Bench | — | 31.8% |
Multilingual Command A leads
Command A: 45.3 (#170), Llama 4 Maverick: 42.2 (#195)
| Benchmark | Command A | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1313 | 1269 |
| LMArena Chinese | 1327 | 1277 |
| LMArena French | 1351 | 1259 |
| LMArena German | 1341 | 1291 |
| LMArena Japanese | 1285 | 1207 |
| LMArena Korean | 1285 | 1203 |
| LMArena Russian | 1314 | 1286 |
| LMArena Spanish | 1347 | 1293 |
Instruction Following Llama 4 Maverick leads
Command A: 69.1 (#177), Llama 4 Maverick: 71.7 (#146)
| Benchmark | Command A | Llama 4 Maverick |
|---|---|---|
| LMArena Instruction Following | 1309 | 1267 |
| IFEval | — | 90.8% |
Long Context Command A leads
Command A: 40.6 (#151), Llama 4 Maverick: 31.4 (#279)
| Benchmark | Command A | Llama 4 Maverick |
|---|---|---|
| LMArena Longer Query | 1334 | 1280 |
| Fiction.LiveBench | — | 46.2% |
Writing & Preference Command A leads
Command A: 47.6 (#208), Llama 4 Maverick: 38.8 (#252)
| Benchmark | Command A | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1331 | 1287 |
| LMArena Creative Writing | 1319 | 1267 |
| EQ-Bench Creative Writing | 1145 | 860 |
| LMArena Multi-Turn | 1339 | 1289 |
| Short-Story Creative Writing | — | 62% |
| WildBench | — | 80% |
Frequently asked questions
Is Command A better than Llama 4 Maverick?
Command A is the stronger model overall, scoring 36.5 to 30.9 on the Noometry Index. Llama 4 Maverick costs 14× less per token, which makes it the better buy when Command A's lead doesn't matter for your workload.
Which is cheaper, Command A or Llama 4 Maverick?
Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; Command A lists at $2.50 and $10.
Is Command A or Llama 4 Maverick better for coding?
They score almost the same on coding (27.2 vs 26.6); test both on your own repository before choosing.
Which has the bigger context window?
Command A does, with 256K tokens against 128K.
How many benchmarks do Command A and Llama 4 Maverick share?
24 benchmarks have published results for both models. Command A has 24 scored results on Noometry and Llama 4 Maverick has 54.