Model comparison
Command A vs GPT-4o
Command A is the stronger model overall, scoring 35.9 to 28.5 on the Noometry Index.
Last verified . 21 shared benchmarks.
CohereSummary
- They share 21 benchmarks with published results for both. Command A scores higher in 8 categories and GPT-4o in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Command A leads 33.5 to 10.2.
- The biggest single-benchmark swing is Aider Polyglot: 12% for Command A and 45.3% for GPT-4o.
- Both cost about the same: $2.50 input and $10 output per million tokens.
- Command A accepts more context: 256K tokens versus 128K.
- Command A has downloadable open weights; the other is API-only.
Side by side
| Command A | GPT-4o | |
|---|---|---|
| Provider | Cohere | OpenAI |
| Noometry Index | 35.9 | 28.5 |
| Released | 2025-03-13 | 2024-05-13 |
| Weights | Open | Proprietary |
| Context window | 256K | 128K |
| Max output | 8K | 16K |
| Input $ / M tokens | $2.50 | $2.50 |
| Output $ / M tokens | $10 | $10 |
| Results tracked | 24 | 72 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Command A leads
Command A: 26.8 (#317), GPT-4o: 25.1 (#323)
| Benchmark | Command A | GPT-4o |
|---|---|---|
| Aider Polyglot | 12% | 45.3% |
| LMArena Coding | 1331 | 1297 |
| SWE-bench Verified | — | 31% |
| SWE-bench Verified (bash only) | — | 21.6% |
| GSO | — | 0% |
| WeirdML | — | 25.1% |
| BigCodeBench Instruct | — | 51.1% |
| LiveBench Coding | — | 51.4% |
| BigCodeBench Complete | — | 61.1% |
| CadEval | — | 26% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 72.2% |
Agentic & Tool Use Command A leads
Command A: 35.7 (#36), GPT-4o: 20.8 (#133)
| Benchmark | Command A | GPT-4o |
|---|---|---|
| Berkeley Function Calling Leaderboard | 57.1% | — |
| GDPval | — | 9.9% |
| TheAgentCompany | — | 8.6% |
| Cybench | — | 12.5% |
| BALROG | — | 32.3% |
| LMArena Search | — | 1006 |
| METR Time Horizons | — | 40.8% |
Reasoning Command A leads
Command A: 17.7 (#276), GPT-4o: 9.4 (#338)
| Benchmark | Command A | GPT-4o |
|---|---|---|
| LMArena Hard Prompts | 1326 | 1281 |
| DTBench | 61.3% | 64.5% |
| LMCA | 10.3% | 16.6% |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 17.8% |
| Kagi LLM Benchmark | 28.8% | — |
| ARC-AGI-1 | — | 4.5% |
| CritPt | — | 0% |
| Chess Puzzles | — | 13% |
| EnigmaEval | — | 0.8% |
| LiveBench Reasoning | — | 55.8% |
| LiveBench Data Analysis | — | 60.9% |
| Epoch Capabilities Index | — | 128.97 |
| ForecastBench | — | 57.7 |
| LiveBench | — | 55.3% |
Math Command A leads
Command A: 33.5 (#172), GPT-4o: 10.2 (#303)
| Benchmark | Command A | GPT-4o |
|---|---|---|
| LMArena Math | 1300 | 1285 |
| FrontierMath (Tiers 1-3) | — | 0.4% |
| OTIS Mock AIME 2024-2025 | — | 6.4% |
| Omni-MATH | — | 29.3% |
| LiveBench Math | — | 49.5% |
| MATH Level 5 | — | 53.3% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Command A leads
Command A: 36.5 (#157), GPT-4o: 28.8 (#235)
| Benchmark | Command A | GPT-4o |
|---|---|---|
| Vectara Hallucination Rate | 9.3% | 9.6% |
| LMArena Expert | 1295 | 1250 |
| GPQA Diamond | — | 49.2% |
| Humanity's Last Exam | — | 2.7% |
| SimpleQA Verified | — | 26% |
| MMLU-Pro | — | 71.3% |
| Confabulations | — | 15.3% |
| GPQA (HELM) | — | 52% |
| MMLU | — | 88.1% |
Multimodal Not comparable
Command A: —, GPT-4o: 33.9 (#85)
| Benchmark | Command A | GPT-4o |
|---|---|---|
| LMArena Vision | — | 1137 |
| Video-MME | — | 71.9% |
| GeoBench | — | 71% |
| VPCT | — | 40% |
| ScienceQA | — | 88.5% |
Multilingual Command A leads
Command A: 45.5 (#166), GPT-4o: 43.4 (#181)
| Benchmark | Command A | GPT-4o |
|---|---|---|
| LMArena Non-English | 1313 | 1283 |
| LMArena Chinese | 1327 | 1277 |
| LMArena French | 1352 | 1303 |
| LMArena German | 1341 | 1282 |
| LMArena Japanese | 1283 | 1257 |
| LMArena Korean | 1285 | 1234 |
| LMArena Russian | 1314 | 1286 |
| LMArena Spanish | 1346 | 1292 |
Instruction Following Command A leads
Command A: 69.3 (#172), GPT-4o: 66.8 (#202)
| Benchmark | Command A | GPT-4o |
|---|---|---|
| LMArena Instruction Following | 1309 | 1278 |
| LiveBench Instruction Following | — | 68.6% |
| IFEval | — | 81.7% |
Long Context Command A leads
Command A: 40.6 (#147), GPT-4o: 39.4 (#174)
| Benchmark | Command A | GPT-4o |
|---|---|---|
| LMArena Longer Query | 1334 | 1289 |
| Fiction.LiveBench | — | 66.7% |
Writing & Preference GPT-4o leads
Command A: 47.8 (#204), GPT-4o: 52.8 (#160)
| Benchmark | Command A | GPT-4o |
|---|---|---|
| LMArena Text | 1331 | 1300 |
| LMArena Creative Writing | 1319 | 1291 |
| LMArena Multi-Turn | 1338 | 1302 |
| Short-Story Creative Writing | — | 81.8% |
| EQ-Bench Creative Writing | 1145 | — |
| WildBench | — | 82.8% |
| LiveBench Language | — | 47.6% |
Frequently asked questions
Is Command A better than GPT-4o?
Command A is the stronger model overall, scoring 35.9 to 28.5 on the Noometry Index.
Which is cheaper, Command A or GPT-4o?
GPT-4o is cheaper. It lists at $2.50 per million input tokens and $10 per million output tokens; Command A lists at $2.50 and $10.
Is Command A or GPT-4o better for coding?
Command A scores higher on coding benchmarks: 26.8 versus 25.1 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 GPT-4o share?
21 benchmarks have published results for both models. Command A has 24 scored results on Noometry and GPT-4o has 72.