How to use Ternary Bonsai 27B with Hermes Agent

PrismML documents Hermes Agent as a supported client for its OpenAI compatible server. Start Ternary Bonsai 27B locally, run Hermes's model setup, and choose the local endpoint. Validate a simple tool call before giving the agent a larger task.

Configure the connection

The exact port must match the server you started. The stock command shown on the model page uses MLX LM's default port, 8080.

hermes setup model

# Choose: Custom OpenAI-compatible endpoint
# Base URL: http://127.0.0.1:8080/v1
# API key: any non-empty string
# Model: use the ID returned by GET /v1/models

Validate before real work

Ask the model to call one harmless tool with one required argument.
Check that the response contains a structured tool call rather than plain text that looks like one.
Run a second test that returns tool output to the model.
Record failures by prompt, runtime version, and model version.

Questions people ask

Does local mean Hermes cannot use web tools?

No. The model can run locally while Hermes calls tools that use the network. Local inference and tool network access are separate choices.

Will every Hermes skill work with this model?

Do not assume it will. Skills that need long planning, strict schemas, or many tool turns should be tested against their own success criteria.

Sources

The benchmark numbers on this page describe one checkpoint, runtime, machine, and test shape. Reproduce the test on the hardware and workload you plan to use before making a product decision.