← model directory

cloudflare

meta/llama-3.3-70b-instruct-fp8-fast

cloudflare/meta/llama-3.3-70b-instruct-fp8-fast

↓ runs free on your own hardware

Llama 3.3 70B quantized to fp8 precision, optimized to be faster.

specs & pricing
Type
text
Provider
cloudflare
Model ID
cloudflare/meta/llama-3.3-70b-instruct-fp8-fast
Context window
24K tokens
Self-hostable
Yes — runs on your own GPU
Input price
$0.32 / 1M tokens
Output price
$2.48 / 1M tokens

Cloud price is billed from prepaid credits when a request fails over to the cloud. Open-weight models run free on GPUs you own — the gateway routes to your nodes first.