Model comparison
Command A vs GPT-3.5-turbo
Command A is the stronger model overall, scoring 36.5 to 23.2 on the Noometry Index. GPT-3.5-turbo costs 5.8× less per token, which makes it the better buy when Command A's lead doesn't matter for your workload.
Last verified . 20 shared benchmarks.
CohereSummary
- They share 20 benchmarks with published results for both. Command A scores higher in 8 categories and GPT-3.5-turbo in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Command A leads 36.2 to 6.3.
- The biggest single-benchmark swing is DTBench: 61.3% for Command A and 48.5% for GPT-3.5-turbo.
- GPT-3.5-turbo is cheaper at $0.50 / $1.50 per million input/output tokens, against $2.50 / $10 for Command A.
- Command A accepts more context: 256K tokens versus 16K.
- Command A has downloadable open weights; the other is API-only.
Side by side
| Command A | GPT-3.5-turbo | |
|---|---|---|
| Provider | Cohere | OpenAI |
| Noometry Index | 36.5 | 23.2 |
| Released | 2025-03-13 | 2023-03-01 |
| Weights | Open | Proprietary |
| Context window | 256K | 16K |
| Max output | 8K | 4K |
| Input $ / M tokens | $2.50 | $0.50 |
| Output $ / M tokens | $10 | $1.50 |
| Results tracked | 24 | 44 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Command A leads
Command A: 27.2 (#322), GPT-3.5-turbo: 23.9 (#331)
| Benchmark | Command A | GPT-3.5-turbo |
|---|---|---|
| LMArena Coding | 1330 | 1136 |
| Aider Polyglot | 12% | — |
| WeirdML | — | 3.5% |
| BigCodeBench Instruct | — | 39.1% |
| BigCodeBench Complete | — | 50.6% |
| HumanEval+ | — | 70.7% |
| MBPP+ | — | 69.7% |
Agentic & Tool Use Not comparable
Command A: 35.9 (#40), GPT-3.5-turbo: —
| Benchmark | Command A | GPT-3.5-turbo |
|---|---|---|
| Berkeley Function Calling Leaderboard | 57.1% | — |
| METR Time Horizons | — | 21.5% |
Reasoning Command A leads
Command A: 18.3 (#283), GPT-3.5-turbo: 13.8 (#332)
| Benchmark | Command A | GPT-3.5-turbo |
|---|---|---|
| LMArena Hard Prompts | 1326 | 1108 |
| DTBench | 61.3% | 48.5% |
| LMCA | 10.3% | 9.7% |
| Kagi LLM Benchmark | 28.8% | — |
| Chess Puzzles | — | 0% |
| Mystery Game Puzzles | — | 3% |
| Adversarial NLI | — | 58.1% |
| BIG-Bench Hard | — | 61.6% |
| CommonsenseQA 2.0 | — | 57% |
| Epoch Capabilities Index | — | 118.55 |
| ForecastBench | — | 50.4 |
| WinoGrande | — | 81.6% |
Math Command A leads
Command A: 36.2 (#171), GPT-3.5-turbo: 6.3 (#327)
| Benchmark | Command A | GPT-3.5-turbo |
|---|---|---|
| LMArena Math | 1300 | 1142 |
| FrontierMath (Tiers 1-3) | — | 0% |
| OTIS Mock AIME 2024-2025 | — | 2.2% |
| MATH Level 5 | — | 15.9% |
| GSM8K | — | 57.8% |
Knowledge Command A leads
Command A: 37.1 (#159), GPT-3.5-turbo: 10.0 (#303)
| Benchmark | Command A | GPT-3.5-turbo |
|---|---|---|
| LMArena Expert | 1295 | 1070 |
| GPQA Diamond | — | 28% |
| Vectara Hallucination Rate | 9.3% | — |
| ARC (AI2) Challenge | — | 87.4% |
| BoolQ | — | 87% |
| MMLU | — | 71.4% |
| OpenBookQA | — | 86% |
| TriviaQA | — | 85.8% |
Multilingual Command A leads
Command A: 45.3 (#170), GPT-3.5-turbo: 31.5 (#258)
| Benchmark | Command A | GPT-3.5-turbo |
|---|---|---|
| LMArena Non-English | 1313 | 1108 |
| LMArena Chinese | 1327 | 1075 |
| LMArena French | 1351 | 1118 |
| LMArena German | 1341 | 1090 |
| LMArena Japanese | 1285 | 1043 |
| LMArena Korean | 1285 | 1019 |
| LMArena Russian | 1314 | 1123 |
| LMArena Spanish | 1347 | 1121 |
Instruction Following Command A leads
Command A: 69.1 (#177), GPT-3.5-turbo: 57.9 (#262)
| Benchmark | Command A | GPT-3.5-turbo |
|---|---|---|
| LMArena Instruction Following | 1309 | 1119 |
Long Context Command A leads
Command A: 40.6 (#151), GPT-3.5-turbo: 34.0 (#254)
| Benchmark | Command A | GPT-3.5-turbo |
|---|---|---|
| LMArena Longer Query | 1334 | 1121 |
Writing & Preference Command A leads
Command A: 47.6 (#208), GPT-3.5-turbo: 25.3 (#305)
| Benchmark | Command A | GPT-3.5-turbo |
|---|---|---|
| LMArena Text | 1331 | 1125 |
| LMArena Creative Writing | 1319 | 1092 |
| EQ-Bench Creative Writing | 1145 | 451 |
| LMArena Multi-Turn | 1339 | 1117 |
Frequently asked questions
Is Command A better than GPT-3.5-turbo?
Command A is the stronger model overall, scoring 36.5 to 23.2 on the Noometry Index. GPT-3.5-turbo costs 5.8× 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 GPT-3.5-turbo?
GPT-3.5-turbo is cheaper. It lists at $0.50 per million input tokens and $1.50 per million output tokens; Command A lists at $2.50 and $10.
Is Command A or GPT-3.5-turbo better for coding?
Command A scores higher on coding benchmarks: 27.2 versus 23.9 in the Noometry coding category.
Which has the bigger context window?
Command A does, with 256K tokens against 16K.
How many benchmarks do Command A and GPT-3.5-turbo share?
20 benchmarks have published results for both models. Command A has 24 scored results on Noometry and GPT-3.5-turbo has 44.