Use Oh My Pi with Models Hub: models.yml Custom Provider

Edit Oh My Pi's ~/.omp/agent/models.yml to add Models Hub as a custom provider over the OpenAI-compatible protocol, with verification and error troubleshooting.

Oh My Pi (the omp command) is a terminal AI coding agent with built-in LSP, subagents, and browser tools, configured through ~/.omp/agent/models.yml for custom model providers. Add Models Hub there, and every chat and code edit inside omp uses your Models Hub balance for the model you choose. This page covers the connection only, not how to use Oh My Pi itself.

Three things you need

A Models Hub API key, the base URL https://modelsok.com/v1, and a model name. There's no one-click setup — Oh My Pi isn't one of the targets the Models Hub CLI supports.

Prerequisites

Oh My Pi installed (the omp command available). Use bun install -g @oh-my-pi/pi-coding-agent, or curl, Homebrew, Nix, or PowerShell — see omp.sh for options.
A Models Hub API key, created on the Tokens page. The token's group must have access to the models you plan to use.

Recommended models

Use case	Model	Pricing	Live status
Everyday coding	`claude-sonnet-4-6`	Price	Availability and latency
Harder reasoning and long tasks	`gpt-5.6-luna`	Price	Availability and latency
Cheap and fast for small edits	`gpt-5.4-mini`	Price	Availability and latency

Model names are authoritative on the pricing page.

Setup

Step 1 — Set the environment variable

export MODELSOK_API_KEY="your-models-hub-api-key"

Step 2 — Edit models.yml

Create or edit ~/.omp/agent/models.yml:

providers:
  modelsok:
    baseUrl: https://modelsok.com/v1
    api: openai-completions
    apiKey: MODELSOK_API_KEY
    authHeader: true
    models:
      - id: claude-sonnet-4-6
        name: Claude Sonnet 4.6
        contextWindow: 200000
        maxTokens: 128000
      - id: gpt-5.4-mini
        name: GPT-5.4 Mini
        contextWindow: 128000
        maxTokens: 16384

Key fields

Field	Description
`baseUrl`	The base URL, ending in `/v1`
`api`	Protocol type; `openai-completions` for the OpenAI-compatible protocol — Oh My Pi also supports `anthropic-messages` and others
`apiKey`	The name of an environment variable (no `$` prefix); Oh My Pi looks for that env var first, and only falls back to treating the value as a literal key
`authHeader: true`	Sends the resolved key as `Authorization: Bearer <key>`
`models[].id`	Must exactly match the model name in the Models Hub console

Reasoning models may need a compat override

If a reasoning model returns field names (like the reasoning-content field or the max_tokens parameter name) that differ from the standard OpenAI shape, add a compat block under that model entry to override them — see the official docs/models.md. Everyday usage usually doesn't need this; get the minimal config above working first, then adjust as needed.

Verify

Go to your project directory and launch with the model flag:

cd /path/to/your-project
omp --model modelsok/claude-sonnet-4-6

Or start omp normally, then press Ctrl+P or type /model to switch to a model under modelsok. Send a test message; a normal reply means you're connected. Then open Usage logs in the Models Hub console; a request under your configured model name appears within seconds.

Common errors

Symptom	Cause	Fix
Provider not found on startup	The path or YAML indentation in `~/.omp/agent/models.yml` is wrong	Check the file path, and confirm indentation uses spaces, not tabs
`401 Unauthorized`	`MODELSOK_API_KEY` isn't exported in the current shell, or `apiKey` has a typo	Check the environment variable, reopen the terminal, and retry
Model not found	`models[].id` doesn't match the Models Hub console	Check the name on the pricing page
Tool-call errors with a reasoning model	The reasoning-content field name doesn't match upstream	Override the relevant field in a `compat` block per the official docs
`omp models modelsok` shows nothing	The config wasn't saved, or has a formatting error	Recheck the YAML syntax, then restart `omp`

FAQ

How is it billed? Per input and output token actually consumed, at the price on each model's pricing page; every request is itemized in usage logs.

How is this different from using the official APIs directly? Oh My Pi sends requests to the Models Hub gateway over the openai-completions protocol — you don't need a separate account with each vendor, and one key covers every model.

How do I switch models? Pass --model modelsok/<model-id> at launch, or press Ctrl+P to cycle, or type /model inside a session to open the picker.

CLI one-command setup: Claude Code, Codex, Gemini CLI and OpenClaw support one-click configuration
Model pricing and model status
Pi setup, OpenCode setup

Use Oh My Pi with Models Hub: models.yml Custom Provider

On this page