Decentralized LLM Inference Powered by Solana

Access powerful language models with blazing-fast API and pay-as-you-go pricing on Solana

product

Powerful Features for Modern AI Development

Pay-as-you-go
Simple token-based pricing with Solana. Only pay for what you use.
High Performance
Optimized infrastructure for maximum throughput
Developer Friendly
Simple API with comprehensive documentation
Multiple Models
Access to GPT-4, Claude, Gemini and other leading LLMs through a unified API
chatgpt-beam
gemini-beam
ollama-beam
octora-beam
deepseek-beam
claude-beam
xai-beam

Simple, Usage-Based Pricing

Pay only for what you use, settled instantly on Solana

Example Costs

Based on Claude 3.5 Sonnet model (0.03 SOL/1M input tokens, 0.12 SOL/1M output tokens)

Chat

250 input tokens + 1500 output tokens

= 0.0001875 SOL

Code Completion

100 input tokens + 500 output tokens

= 0.000063 SOL

Context Aware Code Generation

100 000 input tokens + 10 000 output tokens

= 0.0042 SOL