Llama 3.3 70B Instruct

Provider: Meta Llama US

Input price
$0.71
/ 1M tokens
Output price
$0.71
/ 1M tokens
Cache price
$0.71
/ 1M tokens (read)
Context
131K tokens
Max output 115K tokens

Overview

Meta's 70B flagship multilingual text model handles function-calling and json-mode within a 131k context window.

Strengths

  • Symmetric input and output pricing at $0.71 per million tokens
  • Supports 131k context and 115k max output tokens
  • Built-in function calling and structured JSON output modes

Best use cases

  • Complex multilingual instruction-following tasks
  • Large-scale text extraction using JSON mode
  • Multi-turn agentic workflows requiring function calling

Pricing position

Matches Mistral Large 4 input cost of $0.70/M but delivers one-third of its output price.

Caveats

Requires paid hosting with no free tier available, and features a December 2023 knowledge cutoff.

Notes drafted by an LLM from the figures on this page (2026-11-13) and rewritten when the price changes.

Key facts

Price basis
Median third-party host price (no first-party API)
Max output
115K tokens
Knowledge cutoff
2023-12-31
Listed on OpenRouter
2024-12-06
Tracked since
2026-05-08
Last updated
2026-11-10

Features

Function callingJSON output

Monthly cost estimate

Scale Input (M) Output (M) Monthly
0.1M (hobby) 0.1 0.05 $0.11
1M (small app) 1 0.5 $1.06
10M (mid scale) 10 5 $10.65
100M (large) 100 50 $106.50

※ Cache and batch discounts not factored in. Detailed calculator

Official resources

Other models from Meta Llama

Similar-priced models

Source: OpenRouter API (meta-llama/llama-3.3-70b-instruct)