Model comparison
Command A vs Llama 3.1-8B
Command A is the stronger model overall, scoring 36.5 to 23.0 on the Noometry Index. Llama 3.1-8B costs 76× less per token, which makes it the better buy when Command A's lead doesn't matter for your workload.
Last verified . 21 shared benchmarks.
Cohere
MetaSummary
- They share 21 benchmarks with published results for both. Command A scores higher in 9 categories and Llama 3.1-8B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Command A leads 37.1 to 8.0.
- The biggest single-benchmark swing is Berkeley Function Calling Leaderboard: 57.1% for Command A and 25.8% for Llama 3.1-8B.
- Llama 3.1-8B is cheaper at $0.05 / $0.08 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 3.1-8B | |
|---|---|---|
| Provider | Cohere | Meta |
| Noometry Index | 36.5 | 23.0 |
| Released | 2025-03-13 | 2024-07-23 |
| Weights | Open | Open |
| Context window | 256K | 128K |
| Max output | 8K | 4K |
| Input $ / M tokens | $2.50 | $0.05 |
| Output $ / M tokens | $10 | $0.08 |
| Results tracked | 24 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Command A leads
Command A: 27.2 (#322), Llama 3.1-8B: 20.2 (#340)
| Benchmark | Command A | Llama 3.1-8B |
|---|---|---|
| LMArena Coding | 1330 | 1195 |
| Aider Polyglot | 12% | — |
| SciCode | — | 13.2% |
| WeirdML | — | 1.7% |
| BigCodeBench Instruct | — | 32.8% |
| BigCodeBench Complete | — | 40.5% |
| HumanEval+ | — | 62.8% |
| MBPP+ | — | 55.6% |
Agentic & Tool Use Command A leads
Command A: 35.9 (#40), Llama 3.1-8B: 22.5 (#131)
| Benchmark | Command A | Llama 3.1-8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 57.1% | 25.8% |
| BALROG | — | 15.1% |
Reasoning Command A leads
Command A: 18.3 (#283), Llama 3.1-8B: 14.9 (#321)
| Benchmark | Command A | Llama 3.1-8B |
|---|---|---|
| LMArena Hard Prompts | 1326 | 1175 |
| DTBench | 61.3% | 50.9% |
| LMCA | 10.3% | 5.4% |
| Kagi LLM Benchmark | 28.8% | — |
| CritPt | — | 0% |
| Chess Puzzles | — | 0% |
| Epoch Capabilities Index | — | 116.57 |
| PIQA | — | 81.2% |
Math Command A leads
Command A: 36.2 (#171), Llama 3.1-8B: 10.2 (#317)
| Benchmark | Command A | Llama 3.1-8B |
|---|---|---|
| LMArena Math | 1300 | 1179 |
| OTIS Mock AIME 2024-2025 | — | 1.7% |
| Omni-MATH | — | 13.7% |
| MATH Level 5 | — | 22.9% |
| GSM8K | — | 82.4% |
Knowledge Command A leads
Command A: 37.1 (#159), Llama 3.1-8B: 8.0 (#307)
| Benchmark | Command A | Llama 3.1-8B |
|---|---|---|
| LMArena Expert | 1295 | 1144 |
| GPQA Diamond | — | 27% |
| MMLU-Pro | — | 40.6% |
| Vectara Hallucination Rate | 9.3% | — |
| GPQA (HELM) | — | 24.7% |
| BoolQ | — | 82.8% |
| MMLU | — | 56.1% |
Multilingual Command A leads
Command A: 45.3 (#170), Llama 3.1-8B: 34.0 (#249)
| Benchmark | Command A | Llama 3.1-8B |
|---|---|---|
| LMArena Non-English | 1313 | 1148 |
| LMArena Chinese | 1327 | 1151 |
| LMArena French | 1351 | 1177 |
| LMArena German | 1341 | 1144 |
| LMArena Japanese | 1285 | 1061 |
| LMArena Korean | 1285 | 1053 |
| LMArena Russian | 1314 | 1158 |
| LMArena Spanish | 1347 | 1169 |
Instruction Following Command A leads
Command A: 69.1 (#177), Llama 3.1-8B: 58.9 (#258)
| Benchmark | Command A | Llama 3.1-8B |
|---|---|---|
| LMArena Instruction Following | 1309 | 1159 |
| IFEval | — | 74.3% |
Long Context Command A leads
Command A: 40.6 (#151), Llama 3.1-8B: 35.8 (#238)
| Benchmark | Command A | Llama 3.1-8B |
|---|---|---|
| LMArena Longer Query | 1334 | 1182 |
Writing & Preference Command A leads
Command A: 47.6 (#208), Llama 3.1-8B: 29.7 (#290)
| Benchmark | Command A | Llama 3.1-8B |
|---|---|---|
| LMArena Text | 1331 | 1187 |
| LMArena Creative Writing | 1319 | 1154 |
| EQ-Bench Creative Writing | 1145 | 713 |
| LMArena Multi-Turn | 1339 | 1172 |
| WildBench | — | 68.7% |
Frequently asked questions
Is Command A better than Llama 3.1-8B?
Command A is the stronger model overall, scoring 36.5 to 23.0 on the Noometry Index. Llama 3.1-8B costs 76× 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 3.1-8B?
Llama 3.1-8B is cheaper. It lists at $0.05 per million input tokens and $0.08 per million output tokens; Command A lists at $2.50 and $10.
Is Command A or Llama 3.1-8B better for coding?
Command A scores higher on coding benchmarks: 27.2 versus 20.2 in the Noometry coding category.
Which has the bigger context window?
Command A does, with 256K tokens against 128K.
How many benchmarks do Command A and Llama 3.1-8B share?
21 benchmarks have published results for both models. Command A has 24 scored results on Noometry and Llama 3.1-8B has 43.