Model Fleet

Choose Your Intelligence Level

Select the right neural model for your workload. From lightning-fast edge models to our most capable reasoning engine.

Nano-8B

Edge / Fast

Lightning fast responses for simple tasks and extraction.

Context Window8,000 tokens
Generation SpeedVery High (~120 t/s)
Input (per 1M)$0.10
Output (per 1M)$0.20

Capabilities

  • Basic reasoning
  • Text summarization
  • Classification
Most Popular

Proxima-70B

Balanced / Capable

The sweet spot of intelligence, speed, and cost efficiency.

Context Window128,000 tokens
Generation SpeedHigh (~50 t/s)
Input (per 1M)$0.50
Output (per 1M)$1.50

Capabilities

  • Complex coding
  • Creative writing
  • Multilingual

Omni-Max

Heavy / Reasoning

Maximum intelligence for solving the hardest multi-step problems.

Context Window200,000 tokens
Generation SpeedModerate (~20 t/s)
Input (per 1M)$5.00
Output (per 1M)$15.00

Capabilities

  • Advanced math
  • Deep analysis
  • System architecture
Lightswind Pro - Premium React Components & Advanced UI Templates