# Subscription Plans & Features

# Evaluation

• Access to all available models at no cost
• Very limited throughput and no SLA
• 1M tokens for 30 Days
• Access all LLMs at FP16 quality
• Lowest latency, fast responses every time
• Pay-as-you-go inference — tokens only used during runs

# Exploratory

• Billed in CHF per 1 million input and output tokens, by model, at prevailing market rates
• Access all LLMs at FP16 quality
• Lowest latency, fast responses every time
• Pay-as-you-go inference — tokens only used during runs
• 100% Swiss sovereign — no data sharing or leakage
• Auto-renewed monthly

# Standard

• Billed in CHF per 1 million input and output tokens, by model, at prevailing market rates
• Access all LLMs at FP16 quality
• Lowest latency, fast responses every time
• Pay-as-you-go inference — tokens only used during runs
• 100% Swiss sovereign — no data sharing or leakage
• Auto-renewed monthly

# Scale-Up

• Billed in CHF per 1 million input and output tokens, by model, at prevailing market rates
• Access all LLMs at FP16 quality
• Lowest latency, fast responses every time
• Pay-as-you-go inference — tokens only used during runs
• 100% Swiss sovereign — no data sharing or leakage
• Auto-renewed monthly