Command A vs gpt-oss-120b
Benchmarks, API pricing and specs, head to head. Data updated 2026-06-10.
Command A
Cohere · Mar 2025
Cohere's enterprise RAG specialist — 111B dense, two-GPU deployable, with downloadable weights for private deployments.
gpt-oss-120b
OpenAI · Aug 2025
OpenAI's open-weights MoE reasoning model under Apache 2.0 — near o4-mini quality, runnable on a single 80GB GPU.
The verdict
These two models don't yet share verified results on the benchmarks we track, so judge them on specs, pricing and intended use. On price the gap is dramatic: gpt-oss-120b works out roughly 22x cheaper per blended million tokens. Command A also takes 262K of context versus 131K for gpt-oss-120b.
Specs & pricing
| Command A | gpt-oss-120b | |
|---|---|---|
| modhub Index | — | — |
| Input price / 1M | $2.5 | $0.1 |
| Output price / 1M | $10 | $0.5 |
| Context window | 262K | 131K |
| Max output | 8K | 131K |
| Open weights | yes (CC-BY-NC 4.0) | yes (Apache 2.0) |
| Reasoning model | no | yes |
| Multimodal input | text | text |
| Knowledge cutoff | Jun 2024 | Jun 2024 |
| Released | Mar 2025 | Aug 2025 |
| Example monthly cost* | $40.00 | $1.75 |
* 10M input + 1.5M output tokens per month at list prices, no caching. Green = better value on that row.
Frequently asked questions
- Which is better, Command A or gpt-oss-120b?
- These two models don't yet share verified results on the benchmarks we track, so judge them on specs, pricing and intended use. On price the gap is dramatic: gpt-oss-120b works out roughly 22x cheaper per blended million tokens. Command A also takes 262K of context versus 131K for gpt-oss-120b.
- Which is cheaper, Command A or gpt-oss-120b?
- Command A costs $2.5/$10 per million input/output tokens, while gpt-oss-120b costs $0.1/$0.5. For a typical workload of 10M input and 1.5M output tokens per month, that's $40.00 versus $1.75.
- Which model is better for coding, Command A or gpt-oss-120b?
- We don't yet track SWE-bench Verified results for both models; check their individual pages for coding-related scores.