Model comparison
Command A vs Llama-3.3-70B-Instruct
Command A is the stronger model overall, scoring 36.5 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 28× 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 6 categories and Llama-3.3-70B-Instruct in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Command A leads 36.2 to 15.3.
- The biggest single-benchmark swing is Berkeley Function Calling Leaderboard: 57.1% for Command A and 31.9% for Llama-3.3-70B-Instruct.
- Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 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-3.3-70B-Instruct | |
|---|---|---|
| Provider | Cohere | Meta |
| Noometry Index | 36.5 | 30.6 |
| Released | 2025-03-13 | 2024-12-06 |
| Weights | Open | Open |
| Context window | 256K | 128K |
| Max output | 8K | 4K |
| Input $ / M tokens | $2.50 | $0.10 |
| Output $ / M tokens | $10 | $0.32 |
| Results tracked | 24 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama-3.3-70B-Instruct leads
Command A: 27.2 (#322), Llama-3.3-70B-Instruct: 31.0 (#290)
| Benchmark | Command A | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Coding | 1330 | 1268 |
| Aider Polyglot | 12% | — |
| SciCode | — | 26% |
| WeirdML | — | 14.4% |
| BigCodeBench Instruct | — | 46.9% |
| LiveBench Coding | — | 36.6% |
| BigCodeBench Complete | — | 57.5% |
Agentic & Tool Use Command A leads
Command A: 35.9 (#40), Llama-3.3-70B-Instruct: 25.8 (#105)
| Benchmark | Command A | Llama-3.3-70B-Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | 57.1% | 31.9% |
| BALROG | — | 23% |
Reasoning Command A leads
Command A: 18.3 (#283), Llama-3.3-70B-Instruct: 14.1 (#327)
| Benchmark | Command A | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Hard Prompts | 1326 | 1257 |
| DTBench | 61.3% | 59.5% |
| LMCA | 10.3% | 17.5% |
| SimpleBench | — | 19.9% |
| Kagi LLM Benchmark | 28.8% | — |
| CritPt | — | 0% |
| LiveBench Reasoning | — | 50.8% |
| LiveBench Data Analysis | — | 49.5% |
| Epoch Capabilities Index | — | 127.33 |
| ForecastBench | — | 58.6 |
| LiveBench | — | 50.2% |
Math Command A leads
Command A: 36.2 (#171), Llama-3.3-70B-Instruct: 15.3 (#298)
| Benchmark | Command A | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Math | 1300 | 1267 |
| OTIS Mock AIME 2024-2025 | — | 5.1% |
| LiveBench Math | — | 42.2% |
| MATH Level 5 | — | 41.6% |
Knowledge Command A leads
Command A: 37.1 (#159), Llama-3.3-70B-Instruct: 30.6 (#226)
| Benchmark | Command A | Llama-3.3-70B-Instruct |
|---|---|---|
| Vectara Hallucination Rate | 9.3% | 4.1% |
| LMArena Expert | 1295 | 1225 |
| GPQA Diamond | — | 47.4% |
| Confabulations | — | 22.8% |
| MMLU | — | 86.3% |
Multilingual Command A leads
Command A: 45.3 (#170), Llama-3.3-70B-Instruct: 39.9 (#220)
| Benchmark | Command A | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Non-English | 1313 | 1236 |
| LMArena Chinese | 1327 | 1217 |
| LMArena French | 1351 | 1281 |
| LMArena German | 1341 | 1251 |
| LMArena Japanese | 1285 | 1150 |
| LMArena Korean | 1285 | 1143 |
| LMArena Russian | 1314 | 1252 |
| LMArena Spanish | 1347 | 1270 |
Instruction Following Llama-3.3-70B-Instruct leads
Command A: 69.1 (#177), Llama-3.3-70B-Instruct: 71.1 (#157)
| Benchmark | Command A | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Instruction Following | 1309 | 1242 |
| LiveBench Instruction Following | — | 82.7% |
Long Context Command A leads
Command A: 40.6 (#151), Llama-3.3-70B-Instruct: 26.4 (#295)
| Benchmark | Command A | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Longer Query | 1334 | 1256 |
| Fiction.LiveBench | — | 33.3% |
Writing & Preference Too close to call
Command A: 47.6 (#208), Llama-3.3-70B-Instruct: 47.6 (#207)
| Benchmark | Command A | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Text | 1331 | 1274 |
| LMArena Creative Writing | 1319 | 1250 |
| LMArena Multi-Turn | 1339 | 1280 |
| EQ-Bench Creative Writing | 1145 | — |
| LiveBench Language | — | 39.2% |
Frequently asked questions
Is Command A better than Llama-3.3-70B-Instruct?
Command A is the stronger model overall, scoring 36.5 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 28× 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-3.3-70B-Instruct?
Llama-3.3-70B-Instruct is cheaper. It lists at $0.10 per million input tokens and $0.32 per million output tokens; Command A lists at $2.50 and $10.
Is Command A or Llama-3.3-70B-Instruct better for coding?
Llama-3.3-70B-Instruct scores higher on coding benchmarks: 31.0 versus 27.2 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-3.3-70B-Instruct share?
21 benchmarks have published results for both models. Command A has 24 scored results on Noometry and Llama-3.3-70B-Instruct has 43.