v0.2.5 - Provider-Specific Rate Limiting
What's New
Provider-Specific Rate Limiting
Rate limits are now tailored to each embedding provider:
| Provider | Concurrency | Interval | Notes |
|---|---|---|---|
| GitHub Copilot | 1 | 4s | Strict limits (15 req/min) |
| OpenAI | 3 | 500ms | Moderate limits |
| 5 | 200ms | Higher limits | |
| Ollama | 5 | None | Local, no rate limits! |
Use Ollama for Large Codebases
For codebases with 10k+ files, use Ollama locally for fastest indexing:
// .opencode/codebase-index.json
{
"embeddingProvider": "ollama"
}Requires: ollama pull nomic-embed-text
Full Changelog
- feat: provider-specific rate limiting (Ollama has no limits)