Model comparison
Command A vs DeepSeek-V2.5 (Sep 2024)
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.0 to 35.9 on the Noometry Index.
Last verified . 18 shared benchmarks.
Cohere
DeepSeekSummary
- They share 18 benchmarks with published results for both. Command A scores higher in 5 categories and DeepSeek-V2.5 (Sep 2024) in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek-V2.5 (Sep 2024) leads 24.5 to 17.7.
- The biggest single-benchmark swing is Aider Polyglot: 12% for Command A and 17.8% for DeepSeek-V2.5 (Sep 2024).
Side by side
| Command A | DeepSeek-V2.5 (Sep 2024) | |
|---|---|---|
| Provider | Cohere | DeepSeek |
| Noometry Index | 35.9 | 37.0 |
| Released | 2025-03-13 | 2024-09-06 |
| Weights | Open | Open |
| Context window | 256K | — |
| Max output | 8K | — |
| Input $ / M tokens | $2.50 | — |
| Output $ / M tokens | $10 | — |
| Results tracked | 24 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V2.5 (Sep 2024) leads
Command A: 26.8 (#317), DeepSeek-V2.5 (Sep 2024): 31.3 (#274)
| Benchmark | Command A | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| Aider Polyglot | 12% | 17.8% |
| LMArena Coding | 1331 | 1310 |
| BigCodeBench Instruct | — | 48.6% |
| BigCodeBench Complete | — | 53.2% |
| HumanEval+ | — | 83.5% |
| MBPP+ | — | 74.1% |
Agentic & Tool Use Not comparable
Command A: 35.7 (#36), DeepSeek-V2.5 (Sep 2024): —
| Benchmark | Command A | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| Berkeley Function Calling Leaderboard | 57.1% | — |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
Command A: 17.7 (#276), DeepSeek-V2.5 (Sep 2024): 24.5 (#145)
| Benchmark | Command A | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Hard Prompts | 1326 | 1289 |
| Kagi LLM Benchmark | 28.8% | — |
| DTBench | 61.3% | — |
| LMCA | 10.3% | — |
Math Too close to call
Command A: 33.5 (#172), DeepSeek-V2.5 (Sep 2024): 33.3 (#177)
| Benchmark | Command A | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Math | 1300 | 1289 |
Knowledge Command A leads
Command A: 36.5 (#157), DeepSeek-V2.5 (Sep 2024): 34.2 (#190)
| Benchmark | Command A | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Expert | 1295 | 1266 |
| Vectara Hallucination Rate | 9.3% | — |
Multilingual Command A leads
Command A: 45.5 (#166), DeepSeek-V2.5 (Sep 2024): 42.7 (#188)
| Benchmark | Command A | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Non-English | 1313 | 1273 |
| LMArena Chinese | 1327 | 1317 |
| LMArena French | 1352 | 1288 |
| LMArena German | 1341 | 1258 |
| LMArena Japanese | 1283 | 1228 |
| LMArena Korean | 1285 | 1209 |
| LMArena Russian | 1314 | 1289 |
| LMArena Spanish | 1346 | 1248 |
Instruction Following Command A leads
Command A: 69.3 (#172), DeepSeek-V2.5 (Sep 2024): 67.7 (#189)
| Benchmark | Command A | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Instruction Following | 1309 | 1280 |
Long Context Command A leads
Command A: 40.6 (#147), DeepSeek-V2.5 (Sep 2024): 39.5 (#169)
| Benchmark | Command A | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Longer Query | 1334 | 1301 |
Writing & Preference DeepSeek-V2.5 (Sep 2024) leads
Command A: 47.8 (#204), DeepSeek-V2.5 (Sep 2024): 50.0 (#182)
| Benchmark | Command A | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Text | 1331 | 1294 |
| LMArena Creative Writing | 1319 | 1285 |
| LMArena Multi-Turn | 1338 | 1297 |
| EQ-Bench Creative Writing | 1145 | — |
Frequently asked questions
Is Command A better than DeepSeek-V2.5 (Sep 2024)?
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.0 to 35.9 on the Noometry Index.
Is Command A or DeepSeek-V2.5 (Sep 2024) better for coding?
DeepSeek-V2.5 (Sep 2024) scores higher on coding benchmarks: 31.3 versus 26.8 in the Noometry coding category.
How many benchmarks do Command A and DeepSeek-V2.5 (Sep 2024) share?
18 benchmarks have published results for both models. Command A has 24 scored results on Noometry and DeepSeek-V2.5 (Sep 2024) has 22.