Home / Models / Nousresearch / Hermes 3 70B Instruct

Hermes 3 70B Instruct pricing and routing profile

Hermes 3 70B Instruct from Nousresearch lists at $0.70 per million input tokens and $0.70 per million output tokens, with a 131,072-token context window. Prices from OpenRouter's catalogue on 2026-09-28.

Input, per 1M tokens$0.70
Output, per 1M tokens$0.70
Context window131K
Confidence signalNot exposed

Structured output

Cost per 1,000 requests

Request shapeList cost
Classification or routing (600 tokens in, 10 out)$0.427
Chat reply (2,000 in, 400 out)$1.68
Retrieval-augmented answer (8,000 in, 500 out)$5.95

Before prompt caching. Try your own shape in the cost calculator.

Can Gimbix route to it first?

Not as a first-answer model through OpenRouter: no host returns token log probabilities for Hermes 3 70B Instruct. It can still be the model Gimbix falls back to when a cheaper model is unsure.

Cheaper models that can answer first

Models priced below Hermes 3 70B Instruct that expose a confidence signal. Whether they answer your traffic as well is what a Gimbix pilot measures.

ModelInput / output per 1MCheaper on a chat reply
UnslopNemo 12B$0.40 / $0.4043%
Mistral Small 3.1 24B$0.351 / $0.55545%
Cydonia 24B V4.1$0.30 / $0.5052%
DeepSeek V3.2$0.28 / $0.4257%
DeepSeek V3.2 Exp$0.27 / $0.4158%
Llama 4 Maverick$0.188 / $0.65262%

Questions

How much does Hermes 3 70B Instruct cost?

Hermes 3 70B Instruct lists at $0.70 per million input tokens and $0.70 per million output tokens on OpenRouter as of 2026-09-28. A typical chat reply of 2,000 input and 400 output tokens costs about $1.68 per 1,000 requests.

What is the context window of Hermes 3 70B Instruct?

131,072 tokens.

Can a cheaper model replace Hermes 3 70B Instruct?

Sometimes. Gimbix proves it on your own traffic before routing: the cheaper model must match Hermes 3 70B Instruct on one half of your requests and make at most three answers worse on the other.