Skip to content
InferenceHub

Hermes

Nous Research's CLI agent, on the OpenAI Chat wire. Reaches: the open-weight chat models — gpt-oss, GLM, DeepSeek, Kimi, Nemotron.

1

Add the gateway to ~/.hermes/config.yaml

The base URL ends in /v1 (Hermes appends /chat/completions), and the key goes inline as api_key:

# ~/.hermes/config.yaml
model:
  provider: custom
  base_url: https://app.inferencehub.tech/v1
  api_key: sk-prov-live-YOUR_KEY
  default: gpt-oss-120b
  api_mode: chat_completions
  context_length: 131072
2

Run Hermes

hermes

Troubleshooting

If you see…Fix
"context length" errorHermes requires ≥ 64000 and the gateway's /v1/models doesn't advertise one — keep context_length: 131072 in the config.
"model not found"Set default to an open-weight chat id (gpt-oss-120b, zai-glm-5.2, …) — Claude and Codex ids don't serve on this wire. List the live ids from the Models page.