Call Hermes models through the API for chat completions
Send chat completions requests in an OpenAI-compatible format to get model responses
Best for:Developers
AI Infrastructure / API
Nous Research API is an OpenAI-compatible inference service that provides API access to the Hermes model series and some third-party models through Nous Portal.
Website created
2023-03-22
OpenRouter token usage rank
#17
Source: OpenRouter
Monthly unique visitors
2.377M
Source: SimilarWeb

This product is a model inference interface launched by Nous Research. Users must register at portal.nousresearch.com, obtain an API key, and add credits before they can use it. It is called through standard chat completions and completions endpoints, with the base URL set to https://inference-api.nousresearch.com/v1, and supports models such as Hermes-3-Llama-3.1-405B. Its core strengths are an inference system based on MCTS, CoC, and MoA architectures, while also aggregating access to multiple models and providing a Tool Gateway for proxy integration. It is suitable for developers who need to embed Nous models into applications or agents, as well as AI builders seeking a single account to manage multiple models. Compared with pure model hosting services, it places more emphasis on a portal-style unified entry point and tool aggregation.
One-line summary
Nous Research API is an OpenAI-compatible inference service that provides API access to the Hermes model series and some third-party models through Nous Portal.
What people use it for
Users use it to replace OpenAI calls in code to run Hermes models for conversation or generation tasks, and also configure it as the model provider for Hermes Desktop or agent tools to enable automatic switching.
Best for
AI application teams, platform engineers, and infrastructure leads
How it works
Users typically access the product through the API, then call its built-in capabilities around specific tasks to complete their work.
Product type
AI Infrastructure Tool
Pricing
Unknown
The current enhanced batch dataset does not maintain real-time pricing for this product. Please refer to the official website or official documentation.
APIMaster integration
Unknown
Pending verification
Data confidence
Low
Last verified: 2026-08-30
Supports direct calls to Nous Research models using the standard OpenAI format
Generates complex responses through Monte Carlo Tree Search, Chain of Code, and Mixture of Agents
Initially supports the Hermes 3 Llama 70B and DeepHermes 3 8B Preview models
Create keys and view balances through the portal, with an initial $5 in free credits
An inference service endpoint designed specifically for Nous Research models
Send chat completions requests in an OpenAI-compatible format to get model responses
Best for:Developers
Connect the API to Hermes Agent or similar tools to execute long-running tasks
Best for:AI developers
Submit problems that require multi-step reasoning and get outputs based on MCTS/CoC/MoA
Best for:Researchers
After creating a key, top up credits, and consume the corresponding quota when calling the API
Best for:Developers
We do not yet have reliable evidence that this product supports third-party keys, proxies, custom providers, or a custom base URL, so we do not show speculative setup steps or config values.
Discussion summary
Users generally understand Nous Research API as an OpenAI-compatible inference service, mainly used to access the Hermes model series and Agent tools, with support for architectures such as MCTS, CoC, and MoA to improve reasoning capabilities. Discussion focuses on how the API is used with Agent frameworks, free credits and subscription management, and comparisons with other providers. Users also often share hands-on integration experience with local or desktop Agents, as well as the access flow for the portal website.
Users ask about API key applications, model configuration, tool integration, and prompt writing, with discussion focused on Agent memory, skill persistence, and multi-tool calling.
Users discuss free model access, credit consumption, subscription switching, and ways to avoid extra API key management.
Users compare response speed, cost, compatibility, and model quality, and share switching scenarios and real-world experience.
Users report confusion around the signup flow, UI navigation, and account settings, and discuss how to simplify access and avoid credit card requirements.
Users share applications in code generation, research, automation, and browser operations, and discuss the advantages of unified management through the tool gateway.
We currently classify it under "AI Infrastructure / API", and the page description is based on public materials such as the official website and OpenRouter.
The enhanced page for Nous Research API prioritizes the core tasks and use cases that have already been collected, helping you quickly judge whether it matches your current needs.
There is currently no reliable source confirming that the product supports third-party Keys or custom compatible endpoints, so the page status is shown as pending verification.
Also in the AI Infrastructure / API category, and can be used for side-by-side comparison of different task entry points and product formats.
Also in the AI Infrastructure / API category, and can be used for side-by-side comparison of different task entry points and product formats.
Also in the AI Infrastructure / API category, and can be used for side-by-side comparison of different task entry points and product formats.
Sources:Official website
Last verified: 2026-08-30 · Report a correction