Currently listed through:
Model details
Hermes 3 405B Instruct
Hermes 3 405B Instruct is a large generalist language model released by Nous Research and routed through OpenRouter, positioned as a successor to the Hermes 2 line with a clear emphasis on agent-style workloads. The model is offered under the OpenRouter identifier nousresearch/hermes-3-llama-3.1-405b and inherits the broad 131,072-token context window that Nous Research uses for the Hermes 3 family, making it well suited to long document analysis, multi-turn dialogues, and pipelines where preserving coherence across many turns matters. Its prompt interface exposes the standard OpenRouter sampling controls, including temperature, top_p, top_k, and frequency and presence penalties, so developers can shape response style and determinism without learning a bespoke API surface.
According to the supplied catalog and aggregator descriptions, Hermes 3 405B Instruct is framed as a generalist model with concrete advances over Hermes 2 in agentic capabilities, roleplaying, reasoning, and multi-turn conversation, alongside improved long-context coherence. These qualitative strengths make it a practical choice for building tool-using assistants, persona-driven chat experiences, and reasoning-heavy workflows that benefit from a single model handling planning, explanation, and tool orchestration in one pass. The model is text-in and text-out, which keeps integration simple for chat backends, RAG stacks, and agent harnesses that already operate on string prompts. Priced uniformly on OpenRouter at $1.00 per million tokens for both input and output, it offers predictable cost economics for high-volume agent deployments while still delivering the deep context window that long-running autonomous tasks require.
Quick Info
Powered by- Provider
- OpenRouter
- Model key
- nousresearch/hermes-3-llama-3.1-405b
- Release date
- Aug 16, 2024
- Last updated
- Aug 16, 2024
- Knowledge cutoff
- 2023-12-31
- Input modalities
- Output modalities
- Capabilities
Cost
- Input token cost
- $1.00
- Output token cost
- $1.00
Limits
- Output tokens
- 16,384 tokens
- Context window
- 131,072 tokens