Head-to-Head Pricing Benchmark
Side-by-side pricing and context window comparison for production model selection.
Default Recommendation (120M input + 60M output)
Google: Gemma 4 31B 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 | Nous: Hermes 4 70B | Google: Gemma 4 31B |
|---|---|---|
| Developer | nousresearch | |
| Context Window | 131,072 | 262,144 |
| Max Output Tokens | — | — |
| Input Cost / 1M Tokens | $0.1300 | $0.1300 |
| Output Cost / 1M Tokens | $0.4000 | $0.3800 |
| Projected Monthly Cost | $40 | $38 |
| Vision | ❌ No | ✅ Yes |
| Tool Calling | ❌ No | ✅ Yes |
| Structured Output | ✅ Yes | ✅ Yes |
| Reasoning | ✅ Yes | ✅ Yes |
| Web Search | ❌ No | ❌ No |
| Tokenizer | Llama3 | Gemma |
| MMLU Score | N/A | N/A |
Price History
Current Input / 1M
$0.1300
Current Output / 1M
$0.4000
Price History
Current Input / 1M
$0.1300
Current Output / 1M
$0.3800
Performance History
Current TPS
0.00
Current Latency
0ms
Uptime
100.0%
Performance History
Current TPS
0.00
Current Latency
0ms
Uptime
97.8%
Adjust your workload to see projected monthly costs.
Nous: Hermes 4 70B
$40
per month
Google: Gemma 4 31B
$38
per month
Lower costLive Recommendation
Google: Gemma 4 31B is lower-cost at 120M input + 60M output tokens/month.
Continue evaluation with more “A vs B pricing” decision pages.