Providers
Browse all providers on BenchGecko
BenchGecko tracks 268 AI inference providers, from frontier labs to specialized inference platforms.
Provider Types
| Type | Examples | What They Do |
|---|---|---|
| Frontier Labs | OpenAI, Anthropic, Google DeepMind, xAI, Meta | Build and serve their own models |
| Inference Platforms | Together AI, Fireworks, Groq, Cerebras | Host open-weight models with optimized inference |
| Cloud Providers | AWS Bedrock, Azure AI, Google Cloud Vertex | Offer models through cloud infrastructure |
| Specialized | Cohere, AI21 Labs, Mistral AI | Build domain-specific or efficiency-focused models |
Provider Health Dashboard
The BenchGecko homepage displays a live 24-hour provider health dashboard showing uptime status for all major providers. Current status is also available at benchgecko.ai/status.
Cross-Provider Pricing
The same model often costs different amounts depending on the provider. For example, an open-weight model like Llama 4 may be available from 10+ providers at prices varying by 3-5x. The Pricing page shows every provider's price for every model, making cross-provider arbitrage straightforward.
Provider Profiles
Each provider has a dedicated profile page showing:
- All models offered with current pricing
- Company information (headquarters, founded, funding)
- Uptime history
- Unique features (batch API, fine-tuning, function calling support)
Related Pages
- Model Rankings -- all models across all providers
- Pricing Comparison -- cross-provider pricing
- AI Economy -- company financials and valuations
- Compute Hub -- infrastructure behind the providers