Model comparison
Command A vs DeepSeek-R1
DeepSeek-R1 is the stronger model overall, scoring 42.0 to 35.9 on the Noometry Index.
Last verified . 21 shared benchmarks.
Cohere
DeepSeekSummary
- They share 21 benchmarks with published results for both. Command A scores higher in 1 category and DeepSeek-R1 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-R1 leads 45.9 to 26.8.
- The biggest single-benchmark swing is Aider Polyglot: 12% for Command A and 71.4% for DeepSeek-R1.
- DeepSeek-R1 is cheaper at $0.50 / $2.15 per million input/output tokens, against $2.50 / $10 for Command A.
- Command A accepts more context: 256K tokens versus 164K.
- Command A has downloadable open weights; the other is API-only.
Side by side
| Command A | DeepSeek-R1 | |
|---|---|---|
| Provider | Cohere | DeepSeek |
| Noometry Index | 35.9 | 42.0 |
| Released | 2025-03-13 | 2025-01-20 |
| Weights | Open | Proprietary |
| Context window | 256K | 164K |
| Max output | 8K | 64K |
| Input $ / M tokens | $2.50 | $0.50 |
| Output $ / M tokens | $10 | $2.15 |
| Results tracked | 24 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1 leads
Command A: 26.8 (#317), DeepSeek-R1: 45.9 (#63)
| Benchmark | Command A | DeepSeek-R1 |
|---|---|---|
| Aider Polyglot | 12% | 71.4% |
| LMArena Coding | 1331 | 1427 |
| SciCode | — | 35.7% |
| WeirdML | — | 41.6% |
| LiveBench Coding | — | 66.7% |
| ALE-Bench | — | 804.12 |
| AlgoTune | — | 1.7 |
Agentic & Tool Use Command A leads
Command A: 35.7 (#36), DeepSeek-R1: 30.4 (#68)
| Benchmark | Command A | DeepSeek-R1 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 57.1% | — |
| DeepResearch Bench | — | 35.1% |
| BALROG | — | 34.9% |
| METR Time Horizons | — | 53.8% |
Reasoning Too close to call
Command A: 17.7 (#276), DeepSeek-R1: 18.7 (#253)
| Benchmark | Command A | DeepSeek-R1 |
|---|---|---|
| Kagi LLM Benchmark | 28.8% | 69.4% |
| LMArena Hard Prompts | 1326 | 1415 |
| ARC-AGI-2 | — | 1.3% |
| SimpleBench | — | 40.8% |
| ARC-AGI-1 | — | 21.2% |
| CritPt | — | 1.1% |
| LiveBench Reasoning | — | 83.2% |
| DTBench | 61.3% | — |
| LiveBench Data Analysis | — | 69.8% |
| LMCA | 10.3% | — |
| Epoch Capabilities Index | — | 141.3 |
| ForecastBench | — | 60 |
| LiveBench | — | 71.6% |
Math DeepSeek-R1 leads
Command A: 33.5 (#172), DeepSeek-R1: 42.0 (#70)
| Benchmark | Command A | DeepSeek-R1 |
|---|---|---|
| LMArena Math | 1300 | 1401 |
| OTIS Mock AIME 2024-2025 | — | 66.4% |
| Omni-MATH | — | 42.4% |
| LiveBench Math | — | 80.7% |
| MATH Level 5 | — | 96.6% |
Knowledge DeepSeek-R1 leads
Command A: 36.5 (#157), DeepSeek-R1: 44.3 (#81)
| Benchmark | Command A | DeepSeek-R1 |
|---|---|---|
| Vectara Hallucination Rate | 9.3% | 11.3% |
| LMArena Expert | 1295 | 1394 |
| GPQA Diamond | — | 76.3% |
| MMLU-Pro | — | 79.3% |
| Confabulations | — | 12.7% |
| GPQA (HELM) | — | 66.6% |
Multilingual DeepSeek-R1 leads
Command A: 45.5 (#166), DeepSeek-R1: 52.6 (#80)
| Benchmark | Command A | DeepSeek-R1 |
|---|---|---|
| LMArena Non-English | 1313 | 1412 |
| LMArena Chinese | 1327 | 1443 |
| LMArena French | 1352 | 1415 |
| LMArena German | 1341 | 1405 |
| LMArena Japanese | 1283 | 1394 |
| LMArena Korean | 1285 | 1359 |
| LMArena Russian | 1314 | 1423 |
| LMArena Spanish | 1346 | 1410 |
Instruction Following DeepSeek-R1 leads
Command A: 69.3 (#172), DeepSeek-R1: 72.2 (#138)
| Benchmark | Command A | DeepSeek-R1 |
|---|---|---|
| LMArena Instruction Following | 1309 | 1382 |
| LiveBench Instruction Following | — | 80.5% |
| IFEval | — | 78.4% |
Long Context DeepSeek-R1 leads
Command A: 40.6 (#147), DeepSeek-R1: 45.4 (#36)
| Benchmark | Command A | DeepSeek-R1 |
|---|---|---|
| LMArena Longer Query | 1334 | 1391 |
| Fiction.LiveBench | — | 75% |
Writing & Preference DeepSeek-R1 leads
Command A: 47.8 (#204), DeepSeek-R1: 61.5 (#84)
| Benchmark | Command A | DeepSeek-R1 |
|---|---|---|
| LMArena Text | 1331 | 1428 |
| LMArena Creative Writing | 1319 | 1406 |
| EQ-Bench Creative Writing | 1145 | 1500 |
| LMArena Multi-Turn | 1338 | 1404 |
| Short-Story Creative Writing | — | 83% |
| WildBench | — | 82.8% |
| LiveBench Language | — | 48.5% |
Frequently asked questions
Is Command A better than DeepSeek-R1?
DeepSeek-R1 is the stronger model overall, scoring 42.0 to 35.9 on the Noometry Index.
Which is cheaper, Command A or DeepSeek-R1?
DeepSeek-R1 is cheaper. It lists at $0.50 per million input tokens and $2.15 per million output tokens; Command A lists at $2.50 and $10.
Is Command A or DeepSeek-R1 better for coding?
DeepSeek-R1 scores higher on coding benchmarks: 45.9 versus 26.8 in the Noometry coding category.
Which has the bigger context window?
Command A does, with 256K tokens against 164K.
How many benchmarks do Command A and DeepSeek-R1 share?
21 benchmarks have published results for both models. Command A has 24 scored results on Noometry and DeepSeek-R1 has 52.