Head-to-Head Pricing Benchmark
Side-by-side pricing and context window comparison for production model selection.
Default Recommendation (120M input + 60M output)
Cohere: Command R7B (12-2024) is lower-cost for the default monthly workload scenario.
Adjust the workload in the calculator below to see a live recommendation for your usage.
| Metric | Cohere: Command R7B (12-2024) | NVIDIA: Nemotron Nano 9B V2 |
|---|---|---|
| Developer | cohere | nvidia |
| Context Window | 128,000 | 131,072 |
| Max Output Tokens | 4,000 | — |
| Input Cost / 1M Tokens | $0.0375 | $0.0400 |
| Output Cost / 1M Tokens | $0.1500 | $0.1600 |
| Projected Monthly Cost | $14 | $14 |
| Vision | ❌ No | ❌ No |
| Tool Calling | ❌ No | ✅ Yes |
| Structured Output | ✅ Yes | ✅ Yes |
| Reasoning | ❌ No | ✅ Yes |
| Web Search | ❌ No | ❌ No |
| Tokenizer | Cohere | Other |
| MMLU Score | N/A | 73.9 |
| GPQA | N/A | 55.7 |
Price History
Current Input / 1M
$0.0375
Current Output / 1M
$0.1500
Price History
Current Input / 1M
$0.0400
Current Output / 1M
$0.1600
Performance History
Current TPS
0.00
Current Latency
0ms
Uptime
100.0%
Performance History
Current TPS
0.00
Current Latency
0ms
Uptime
100.0%
Adjust your workload to see projected monthly costs.
Cohere: Command R7B (12-2024)
$14
per month
Lower costNVIDIA: Nemotron Nano 9B V2
$14
per month
Live Recommendation
Cohere: Command R7B (12-2024) is lower-cost at 120M input + 60M output tokens/month.
Continue evaluation with more “A vs B pricing” decision pages.