o3 with OpenCode via LLM Gateway

Model Overview

• Model ID: o3
• Provider: OpenAI
• Strengths: Advanced multi-step reasoning, complex problem solving
• Context: 128K tokens

Configuration

Add to ~/.config/opencode/config.json:

{
  "provider": "openai-compatible",
  "apiKey": "your-llmgateway-api-key",
  "baseUrl": "https://api.llmgateway.io/v1",
  "model": "o3"
}

Or via environment:

export OPENCODE_PROVIDER="openai-compatible"
export OPENCODE_API_KEY="your-llmgateway-api-key"
export OPENCODE_BASE_URL="https://api.llmgateway.io/v1"
export OPENCODE_MODEL="o3"

Usage

opencode --model o3 "optimize this algorithm for O(n) complexity"

Best Use Cases

• Algorithm design
• Complex debugging with multiple dependencies
• Mathematical/logical code problems
• Performance optimization analysis
• System design decisions

Notes

• Responses may take longer due to extended reasoning
• Consider increasing timeout in config: "timeout": 120000

Pricing Tier

Premium - use for problems requiring deep reasoning.