Very long inputs
Choose Command A
Command A reads up to 256k tokens per request, against 16k for davinci-002.
Command A vs davinci-002: price, context and capabilities
Choose Command A for very long inputs and the latest release; on the other points the two are level or the data is missing.
In short
- A price comparison is not possible: one of them has no comparable published price.
- Command A reads more per request: 256k tokens against 16k.
- Command A is the newer model (March 2025, against August 2023).
- Only Command A offers tool calling.
- Only davinci-002 offers batch api.
Which should you choose?
Each answer below comes straight from the published prices, specifications and independent benchmark results.
Latest release
Choose Command A
Command A is the newer release (March 2025, against August 2023 for davinci-002).
Specifications side by side
| Command A | davinci-002 | |
|---|---|---|
| Provider | Cohere | OpenAI |
| Category | Language models | Language models |
| Status | Generally available | Deprecated |
| Headline price | No data | No data |
| Input | No data | No data |
| Output | No data | No data |
| Cached input | No data | No data |
| Context window | 256k | 16k |
| Max output | No data | No data |
| Released | March 2025 | August 2023 |
| Knowledge cutoff | No data | September 2021 |
| Licence | Proprietary | Proprietary |
| Capabilities | Tool calling, Structured output | Structured output, Batch API |
| API model id | command-a-03-2025 | davinci-002 |
Benchmarks side by side
Leader in bold.
| Benchmark | Command A | davinci-002 |
|---|---|---|
| Terminal-Bench Hard | 0.8% | No data |
| GPQA Diamond | 52.7% | No data |
| Humanity's Last Exam | 4% | No data |
| IFBench (instruction following) | 36.5% | No data |
| τ²-Bench (agentic tool use) | 15.2% | No data |
| Long-context reasoning (AA-LCR) | 21.3% | No data |
Independent evaluations by Artificial Analysis, run the same way for every model, at the reasoning setting shown. Percentages unless marked as an index.
Frequently asked questions
Should I choose Command A or davinci-002?
Choose Command A for very long inputs and the latest release; on the other points the two are level or the data is missing.
Which is cheaper, Command A or davinci-002?
A price comparison is not possible: one of them has no comparable published price.
Which has the larger context window, Command A or davinci-002?
Command A reads more per request: 256k tokens against 16k.
What is the difference between Command A and davinci-002?
Command A is made by Cohere and davinci-002 by OpenAI. A price comparison is not possible: one of them has no comparable published price. Command A reads more per request: 256k tokens against 16k. Command A is the newer model (March 2025, against August 2023). Only Command A offers tool calling. Only davinci-002 offers batch api.
Which is newer, Command A or davinci-002?
Command A was released on March 2025 and davinci-002 on August 2023.