Skip to main content
OpenCode is an open-source AI coding assistant that runs in your terminal. Built by Anomaly, it provides an intuitive interface for AI-powered code generation and editing.

​
Installation

Install OpenCode from the official source: 
curl -fsSL https://opencode.ai/install | bash
Learn more at opencode.ai.

​
Quick Setup

ollama launch opencode
Ollama automatically:
1

Selects models

Interactive multi-select picker for models
2

Updates configuration

Writes to ~/.config/opencode/opencode.json
3

Updates recents

Adds models to recent list in ~/.local/state/opencode/model.json
4

Launches OpenCode

Starts with configured models available

​
Configuration Only

ollama launch opencode --config

​
Use Specific Model

ollama launch opencode --model qwen3-coder:480b-cloud
OpenCode requires a large context window (at least 64k tokens). See Context Length for configuration.

​
Features

Multiple Models

Select and switch between multiple configured models

Recent History

Quick access to recently used models

Favorites

Bookmark your preferred models

Cloud Support

Automatic configuration for cloud models

​
Cloud Models

glm-4.7:cloud

Recommended model for OpenCode (200k context)
Other excellent options:
  • qwen3-coder:480b-cloud — Advanced code generation (260k context)
  • minimax-m2.5:cloud — Fast, efficient coding (200k context)
  • deepseek-v3.1:671b-cloud — Massive reasoning (160k context)
Explore more at ollama.com/search?c=cloud.

​
Local Models

  • qwen3-coder — Efficient code generation (~11GB VRAM)
  • glm-4.7 — Reasoning and coding (~25GB VRAM)
  • deepseek-coder — Specialized code model (~20GB VRAM)

​
Manual Setup

Add a configuration block to ~/.config/opencode/opencode.json:
{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "ollama": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Ollama",
      "options": {
        "baseURL": "http://localhost:11434/v1"
      },
      "models": {
        "qwen3-coder": {
          "name": "qwen3-coder"
        }
      }
    }
  }
}

​
Cloud Models Configuration

For cloud models with context/output limits:
{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "ollama": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Ollama",
      "options": {
        "baseURL": "http://localhost:11434/v1"
      },
      "models": {
        "glm-4.7:cloud": {
          "name": "glm-4.7:cloud",
          "limit": {
            "context": 202752,
            "output": 131072
          }
        }
      }
    }
  }
}
Ollama automatically adds these limits when you use ollama launch opencode.

​
Configuration Files

OpenCode uses two configuration files:

​
Main Config

~/.config/opencode/opencode.json — Provider and model configuration

​
State File

~/.local/state/opencode/model.json — Recent and favorite models 
{
  "recent": [
    {
      "providerID": "ollama",
      "modelID": "glm-4.7:cloud"
    },
    {
      "providerID": "ollama",
      "modelID": "qwen3-coder"
    }
  ],
  "favorite": [
    {
      "providerID": "ollama",
      "modelID": "glm-4.7:cloud"
    }
  ],
  "variant": {}
}
Ollama automatically manages both files when you use ollama launch opencode.

​
Multiple Models

OpenCode supports multiple models simultaneously. Use ollama launch opencode to configure several models: 
 ollama launch opencode
# → Select multiple models in the picker
# → All models become available in OpenCode
Switch between models in the OpenCode UI.

​
Connecting to ollama.com

To use cloud models hosted on ollama.com:
1

Create an API key

Go to ollama.com/settings/keys
2

Export the key

export OLLAMA_API_KEY=your-key-here
3

Update config

Edit ~/.config/opencode/opencode.json:
{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "ollama": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Ollama Cloud",
      "options": {
        "baseURL": "https://ollama.com/v1",
        "headers": {
          "Authorization": "Bearer ${OLLAMA_API_KEY}"
        }
      },
      "models": {
        "glm-4.7:cloud": {
          "name": "glm-4.7:cloud"
        }
      }
    }
  }
}
4

Restart OpenCode

opencode

​
Usage Examples

​
Start in a Project

cd ~/projects/my-app
opencode
OpenCode opens in your terminal with the configured models available.

​
Switch Models

Use the model picker in the OpenCode UI (usually Ctrl+M or Cmd+M).

​
Access Recent Models

Your most recently used models appear at the top of the picker.

​
Troubleshooting

​
Model Not Available

Ensure the model is pulled: 
ollama pull glm-4.7:cloud
ollama list

​
Configuration Not Loading

Restart OpenCode to pick up config changes: 
# Kill existing OpenCode instances
pkill opencode

# Start fresh
opencode

​
Context Window Issues

For local models, increase context: 
ollama run qwen3-coder /set parameter num_ctx 65536
See Context Length for details.

​
Ollama Connection Failed

Verify Ollama is running: 
ollama list
Check the base URL in your config matches your Ollama host.

​
Advanced Configuration

​
Custom Provider Name

{
  "provider": {
    "ollama": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "My Local Ollama",
      "options": {
        "baseURL": "http://localhost:11434/v1"
      }
    }
  }
}

​
Per-Model Settings

{
  "provider": {
    "ollama": {
      "models": {
        "qwen3-coder": {
          "name": "Qwen3 Coder (Local)",
          "limit": {
            "context": 32768,
            "output": 8192
          }
        }
      }
    }
  }
}

​
Environment Variables

OpenCode respects:
  • OLLAMA_HOST — Override Ollama server URL
  • OLLAMA_API_KEY — API key for ollama.com

​
Backup Configuration

When using ollama launch opencode, Ollama creates backups in ~/.ollama/backups/ before modifying your configuration.

​
Learn More

OpenCode Website

Official OpenCode website

Install Guide

Detailed installation instructions

OpenAI API

Ollama’s OpenAI-compatible API

Context Length

Configure model context windows

Build docs developers (and LLMs) love

Get started for freeTalk to us