Pricing
Free during beta. Full access to inference and smart routing, no credit card required.
Public Beta - No Payment Required
All features including inference are free during beta. Verify your email and get credits to run models. No credit card needed.
Free
BetaRun open models through one API with smart routing.
- Inference API access (beta credits included)
- Model catalog with structured data
- Side-by-side model comparison
- GPU capacity planner
- Model recommender
- Basic saved history
Pro
BetaHigher limits, exports, and priority access to new features.
- Everything in Free
- Higher inference credit allocation
- Unlimited saved models and comparisons
- Export comparisons to CSV and PDF
- Priority access to new models
- Request custom usage limits
Enterprise
For teams adopting AI models at scale.
- Dedicated inference capacity
- Team access and collaboration
- Custom model onboarding
- Dedicated support
- SLA and uptime guarantees
What every account gets
Available to all users. Tier-specific limits apply to saved items and exports.
Inference
- OpenAI-compatible API
Single endpoint, works with existing SDKs
- Smart routing
Automatic model selection based on prompt complexity
- Playground
Test and compare models from your browser
Model selection
- Model catalog
Structured data on hundreds of models, updated weekly
- Side-by-side comparison
Compare models across capabilities and benchmarks
- GPU capacity planner
VRAM calculations and GPU recommendations for self-hosting
Standard rate limits apply to ensure platform stability. If you need higher limits, contact us.
Frequently asked questions
During beta, every account receives free inference credits once its email is verified. Paid plans will come after beta; we will share details and give existing users notice before anything changes.
No. Everything is free during beta, no payment details required.
After beta, once billing and enterprise features are in place. Existing users will get advance notice and a clear migration path.
Yes. You can use the OpenAI SDK or any compatible client. Just point it to our endpoint and use your Inferbase API key.
Team features are planned. For early enterprise access, contact us and we will work out the details.
Start building with the right model.
Automatically route workloads to the right model for every task, every time.