OpenCode Integration

OGX includes built-in support for connecting OpenCode, a terminal-based AI coding assistant, with a single command. All available models on your OGX server are automatically configured and ready to use.

Quick Start

1. Start OGX

export OPENAI_API_KEY="your-key-here"
ogx run starter

2. Connect OpenCode

ogx connect opencode

That's it. OpenCode starts with all your OGX models available and the first model pre-selected.

How It Works

The ogx connect opencode command:

1. Queries the running OGX server for available models
2. Filters out non-LLM models (embeddings, rerankers)
3. Generates an OpenCode provider configuration with all discovered models
4. Launches OpenCode with the configuration via the OPENCODE_CONFIG_CONTENT environment variable

ogx connect opencode
       |
       v
  GET /v1/models (discover available models)
       |
       v
  Generate OpenCode config (all LLM models)
       |
       v
  Launch OpenCode with OPENCODE_CONFIG_CONTENT env var
       |
       v
  OpenCode TUI (connected to OGX)

CLI Reference

ogx connect opencode [--model MODEL] [--host HOST] [--port PORT]

Options:

Flag	Default	Description
--model	First available model	Model to pre-select as the default in OpenCode
--host	localhost	OGX server host
--port	8321	OGX server port (also reads OGX_PORT env var)

All discovered LLM models are available in OpenCode regardless of which model is set as the default. You can switch models inside OpenCode at any time.

Configuration Examples

With a specific default model

ogx connect opencode --model openai/gpt-4o

With a remote OGX server

ogx connect opencode --host 192.168.1.100 --port 9000

With Ollama backend

# Terminal 1: Start OGX with Ollama
export OLLAMA_URL="http://localhost:11434/v1"
ogx run starter

# Terminal 2: Connect OpenCode
ogx connect opencode

With vLLM backend

# Terminal 1: Start OGX with vLLM
export VLLM_URL="http://localhost:8000/v1"
ogx run starter

# Terminal 2: Connect OpenCode
ogx connect opencode --model vllm/Qwen/Qwen3-8B

Prerequisites

• OpenCode must be installed and available in your PATH. Install it from opencode.ai.
• OGX server must be running before connecting OpenCode. Start it with ogx stack run or ogx stack letsgo.

Troubleshooting

"Failed to find 'opencode' in PATH"

Install OpenCode following the instructions at opencode.ai.

"Failed to connect to OGX server"

The OGX server is not running or not reachable at the specified host and port. Start it first:

ogx run starter

"Failed to find any LLM models"

The OGX server is running but has no LLM models registered. Check your distribution configuration and ensure at least one inference provider is configured.

Model not found with --model

The specified model ID must match exactly what the OGX server reports. List available models:

curl http://localhost:8321/v1/models | python -m json.tool

---

For more information about OGX's provider architecture, see Providers Overview. For OpenCode documentation, visit opencode.ai/docs.