Model comparison
Command A vs Qwen2.5-Coder-32B
Command A is the stronger model overall, scoring 36.5 to 33.4 on the Noometry Index. Qwen2.5-Coder-32B costs 5.9× less per token, which makes it the better buy when Command A's lead doesn't matter for your workload.
Last verified . 13 shared benchmarks.
Cohere
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. Command A scores higher in 7 categories and Qwen2.5-Coder-32B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Command A leads 69.1 to 61.4.
- Qwen2.5-Coder-32B is cheaper at $0.66 / $1 per million input/output tokens, against $2.50 / $10 for Command A.
- Command A accepts more context: 256K tokens versus 33K.
Side by side
| Command A | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Cohere | Alibaba (Qwen) |
| Noometry Index | 36.5 | 33.4 |
| Released | 2025-03-13 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 256K | 33K |
| Max output | 8K | 29K |
| Input $ / M tokens | $2.50 | $0.66 |
| Output $ / M tokens | $10 | $1 |
| Results tracked | 24 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Command A leads
Command A: 27.2 (#322), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Command A | Qwen2.5-Coder-32B |
|---|---|---|
| Aider Polyglot | 12% | 16.4% |
| LMArena Coding | 1330 | 1276 |
| SWE-bench Verified (bash only) | — | 9% |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Agentic & Tool Use Not comparable
Command A: 35.9 (#40), Qwen2.5-Coder-32B: —
| Benchmark | Command A | Qwen2.5-Coder-32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 57.1% | — |
Reasoning Qwen2.5-Coder-32B leads
Command A: 18.3 (#283), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Command A | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1326 | 1251 |
| Kagi LLM Benchmark | 28.8% | — |
| LiveBench Reasoning | — | 42.1% |
| DTBench | 61.3% | — |
| LiveBench Data Analysis | — | 49.9% |
| LMCA | 10.3% | — |
| Epoch Capabilities Index | — | 119.49 |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math Command A leads
Command A: 36.2 (#171), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Command A | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1300 | 1251 |
| LiveBench Math | — | 46.6% |
| GSM8K | — | 93% |
Knowledge Command A leads
Command A: 37.1 (#159), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Command A | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1295 | 1221 |
| Vectara Hallucination Rate | 9.3% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multilingual Command A leads
Command A: 45.3 (#170), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Command A | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1313 | 1205 |
| LMArena Chinese | 1327 | 1222 |
| LMArena Russian | 1314 | 1228 |
| LMArena French | 1351 | — |
| LMArena German | 1341 | — |
| LMArena Japanese | 1285 | — |
| LMArena Korean | 1285 | — |
| LMArena Spanish | 1347 | — |
Instruction Following Command A leads
Command A: 69.1 (#177), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Command A | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1309 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Command A leads
Command A: 40.6 (#151), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Command A | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1334 | 1251 |
Writing & Preference Command A leads
Command A: 47.6 (#208), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Command A | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1331 | 1230 |
| LMArena Creative Writing | 1319 | 1174 |
| LMArena Multi-Turn | 1339 | 1222 |
| EQ-Bench Creative Writing | 1145 | — |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Command A better than Qwen2.5-Coder-32B?
Command A is the stronger model overall, scoring 36.5 to 33.4 on the Noometry Index. Qwen2.5-Coder-32B costs 5.9× 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 Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is cheaper. It lists at $0.66 per million input tokens and $1 per million output tokens; Command A lists at $2.50 and $10.
Is Command A or Qwen2.5-Coder-32B better for coding?
Command A scores higher on coding benchmarks: 27.2 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
Command A does, with 256K tokens against 33K.
How many benchmarks do Command A and Qwen2.5-Coder-32B share?
13 benchmarks have published results for both models. Command A has 24 scored results on Noometry and Qwen2.5-Coder-32B has 31.