Free
Vectora’s Free plan is fully open source and time-unlimited. Use it locally on any codebase without paying anything. Perfect for:
- Individual developers
- Small teams (< 10 people)
- Educational projects
- Prototyping and testing
What’s Included
Core Features
| Feature | Free |
|---|---|
| Semantic Search | [x] |
| Dependency Analysis | [x] |
| Full Indexing | [x] |
| Unlimited Namespaces | [x] |
| Trust Folder + Guardian | [x] |
| RBAC (5 roles) | [x] |
| Full CLI | [x] |
| Harness Runtime | [x] |
| Webhooks | [x] |
| SSO/LDAP | [x] |
| SLA Uptime | - |
Integrations
| Platform | Free |
|---|---|
| Claude Code (MCP) | [x] |
| Cursor (MCP) | [x] |
| VS Code Extension | [x] |
| CLI | [x] |
| ChatGPT Custom GPT | (self-hosted) |
| Gemini Integration | (BYOK) |
AI Models (BYOK)
Use your own:
| Component | Option | Cost |
|---|---|---|
| Embedding | Voyage 4 | Free* |
| Reranking | Voyage Rerank 2.5 | Free* |
| LLM | Gemini 3 Flash | Free* |
| Local | Ollama (all-MiniLM) | Free |
*Free tier with limits: 60 req/min (Gemini), 50 req/min (Voyage)
Limits
Rate Limiting
free_tier_limits:
gemini_api:
requests_per_minute: 60
tokens_per_month: 1.5M
voyage_api:
requests_per_minute: 50
tokens_per_month: 100M
vectora_server:
concurrent_users: 1
searches_per_day: 1000Storage
- Vector Index: Unlimited (local disk)
- Embeddings Cache: 500MB
- Logs: 30 days retention
Performance
- Search Latency: <2s
- Concurrent Requests: 1
- Max File Size: 100MB per file
- Max Chunks per Search: 20
How to Start
Installation
# Install globally
npm install -g @kaffyn/vectora
# Verify
vectora --versionFirst Run
# 1. Initialize project
cd ~/your-project
vectora init --name "My Project"
# 2. Obtain free keys
# → Gemini: https://aistudio.google.com/app/apikey
# → Voyage: https://dash.voyageai.com/api-keys
# 3. Configure
vectora config set --key GEMINI_API_KEY
vectora config set --key VOYAGE_API_KEY
# 4. Index
vectora index
# 5. Search!
vectora search "How does login work?"With IDE (Claude Code, Cursor)
// ~/.claude/claude_desktop_config.json
{
"mcpServers": {
"vectora": {
"command": "vectora",
"args": ["mcp"],
"env": {
"GEMINI_API_KEY": "sk-...",
"VOYAGE_API_KEY": "sk-...",
"VECTORA_NAMESPACE": "your-namespace"
}
}
}
}Recommended Use Cases
Best in Free
- Codebase < 100K lines: Indexing in minutes
- 1 project: 1 namespace, isolated Trust Folder
- Teams < 10 people: RBAC available
- Local development: No infrastructure needed
- Education/research: Open source (MIT)
Consider Upgrade
- Multiple large projects: > 5 projects
- Teams > 10 people: Needs LDAP/SSO
- High frequency: > 1000 searches/day
- 24/7 uptime: Needs SLA
- Webhooks: For automated CI/CD
Troubleshooting
“Quota exceeded”
You have reached the monthly API limit (Gemini or Voyage).
Solutions:
- Wait for reset on the 1st of the next month
- Upgrade to Plus (unlimited)
“Rate limit exceeded”
You made too many requests in 1 minute.
Solution: Wait or use spacing:
# Wait between searches
for query in "login" "auth" "token"; do
vectora search "$query"
sleep 1
done“Concurrent users exceeded”
Only 1 user at a time in the free tier (1 IDE).
Solutions:
- Close other instances
- Upgrade to Plus (up to 100 users)
Roadmap to Plus
When you grow:
Free (Your Project)
↓
Plus (Multiple projects + webhooks)
↓
Team (LDAP + SLA + support)
↓
Enterprise (On-premise + support)No migration required - upgrade is automatic.
Support & Community
You’re not alone!
- Docs: vectora.app/docs
- GitHub: github.com/kaffyn/vectora
- Discussions: GitHub Discussions
- Issues: Report bugs
Next: Plus Plan
Part of the Vectora ecosystem · Open Source (MIT) · Contributors