Nous models
4models tracked · ranked by intelligence, speed & price
Match · Updated July 2026
Nous's smartest model is Hermes 3 70B Instruct (5.1 on the Intelligence Index), and its cheapest is Hermes 4 70B at $0.130 per million input tokens. All 4 Nous models are compared below by intelligence, speed, latency, context and price.
Smartest NousFastest NousCheapest NousSmall & fast NousNous for codingNous reasoningNous for designNous for agents
Smartest
5.1
nousresearch/hermes-3-llama-3.1-70b
Fastest
—
Cheapest
$0.130/M
nousresearch/hermes-4-70b
| Model | Intel | Speed | Latency | In $/M | Context |
|---|---|---|---|---|---|
| hermes-3-llama-3.1-70b JSON | 5.1 | — | — | $0.700 | 131K |
| hermes-4-70b ReasoningJSON | — | — | — | $0.130 | 131K |
| hermes-4-405b ReasoningJSON | — | — | — | $1.00 | 131K |
| hermes-3-llama-3.1-405b JSON | — | — | — | $1.00 | 131K |