Responses API · LLM · Endpoint

rellm

A lightweight Go framework for building small, specialized agents on top of the OpenAI Responses API.

Language
Go
Providers
LM Studio · OpenRouter · OpenAI
Output
JSON, schema-validated
Version
v0.1.0
Why small agents

Fewer tools,
fewer mistakes.

Many small, focused agents beat one general-purpose one. A narrow toolset constrains what a model can do; a strict JSON schema constrains what it can say. Local or hosted — your call.

Exhibit I · The idea

Why small, specialized agents?

One agent, one responsibility. Small agents are cheaper to run, easier to measure, and duller to debug — which is exactly what you want.

i.

Fewer tools, fewer mistakes.

A small toolset gives the model fewer ways to pick the wrong tool, so hallucinations have a smaller blast radius.

ii.

Cheaper and faster.

A narrow task needs less model capability — small local models often do the job, so you avoid paying for a frontier model.

iii.

Easy to test.

Success criteria are clear and regressions are easy to spot. A categorization agent's accuracy is a number, not a vibe.

iv.

Composable.

Each agent is a plain Go package — chain them, run them in parallel, or embed one in another agent's toolset.

Exhibit II · The framework

Constrained on both ends.

The model is constrained at both ends — on what it can do and on what it can say. A small, specialized agent becomes very powerful.