All AI models · LLM benchmarks · Methodology

Meta

Llama 3.3 70B Instruct fp8 Fast

Popular open Llama workhorse for multilingual chat, coding, and self-hosting

Model facts

Context window
24K tokens
Maximum output
24K tokens
Cheapest paid input
$0.293 per 1M tokens
Cheapest paid output
$2.25 per 1M tokens
Open weights
yes
Providers
1

Capabilities and modalities

tool calling, temperature control, open weights, text.

Observed price history

  • 2026-08-24: $0.293 input / $2.25 output per 1M tokens

API providers

  • Cloudflare Workers AI — model id @cf/meta/llama-3.3-70b-instruct-fp8-fast; $0.293 input / $2.25 output per 1M tokens; provider documentation