⚡ Lightning Models

AI at Superhuman Speed.

Lightning Text v2 and Lightning Image Turbo are purpose-built for maximum cloud throughput — no compromise on quality.

1,000
Cloud inference
words per second

Two models. Zero compromises.

Text Generation
Lightning Text v2

Built for speed and smooth streaming. Generate instant responses for chat, coding, or long-form writing — with optimized caching and decoding that keeps latency near zero.

Up to 1,000 words/sec streaming
🔄Low-latency optimized decoding
💬Perfect for chat, code & documents
☁️Cloud-native, no local GPU needed
Image Generation
Lightning Image Turbo

Ultra-realistic images with exceptional detail — cinematic, sci-fi, or photorealistic. The new default image model with sharper lighting effects and faster generation than ever before.

🎨Ultra-realistic rendering
🚀Faster than previous models
🌌Exceptional sci-fi & cinematic output
Sharper detail & lighting effects

Lightning is free
for everyone.

Lightning models are available on every plan, including the free tier. Install InferencePort and start generating at 1,000 words/sec today — no credit card needed.

Install InferencePort Free