LLM Pricing

← All Models

Meta: Llama 3.3 70B Instruct

🇺🇸 Meta · Llama 3.3

Input Price — per million tokens
Output Price — per million tokens
Context Window 131K tokens Output limit: 128K
OpenRouter Route Price Please verify with official pricing pages
Use this model via OpenRouter →

Overview

Meta: Llama 3.3 70B Instruct is a large language model API from Meta, part of its Llama 3.3 model family. The 131K-token context window — around 197 pages of text — comfortably handles long documents, multi-file code, or extended conversations. On Artificial Analysis's Intelligence Index it scores 9 (F grade), a useful proxy for its general reasoning strength relative to the other models tracked here. All prices on this page reflect OpenRouter's routed rates and are re-synced automatically every day; confirm against the provider's official pricing before committing to production.

Dimension Unit Price (USD) Price (TWD) Effective From
No pricing data yet

Provider
Meta
Model Family
Llama 3.3
Version String
meta-llama/llama-3.3-70b-instruct
Status
Active
Modality
Text
Context Window
131,072 tokens
Output Limit
128,000 tokens

Index Metrics

Cross-domain capability indexes evaluated by Artificial Analysis — Artificial Analysis

Agentic Index 0 F Measured: 2026-07-25
Coding Index 12 F Measured: 2026-07-25
Intelligence Index 9 F Measured: 2026-07-25

Benchmark Scores

Data source: Artificial Analysis

AA-LCR 15.0% F Measured: 2026-07-25
GPQA Diamond 49.8% C Measured: 2026-07-25
HLE 4.0% D Measured: 2026-07-25
IFBench 47.1% C Measured: 2026-07-25
Non-Hallucination 14.9% Measured: 2026-07-25
Omniscience Accuracy 17.9% Measured: 2026-07-25
SciCode 26.0% C Measured: 2026-07-25
Tau2 26.6% Measured: 2026-07-25
TerminalBench 3.0% Measured: 2026-07-25

Performance Metrics

Real-world benchmarks, updated every 72 hours by Artificial Analysis — Artificial Analysis

First Token Latency 1.8s Measured: 2026-07-25
Output Speed 81 t/s Measured: 2026-07-25
Response Time 7.9s Measured: 2026-07-25

Use Case Analysis

Model strengths, weaknesses, and ideal use cases based on benchmarks and official documentation

Strength

  • 405B-level at 70B Cost MMLU (CoT) 86.0% matches 405B, HumanEval 88.4% exceeds 405B, MATH 77.0% exceeds 405B. Benchmark
  • Superior Instruction Following IFEval 92.1%, best-in-class, even better than 405B (88.6%). Benchmark

Best For

  • Production Enterprise Applications Best value for large-scale deployments — 405B quality at 70B cost and latency. Official
  • Instruction-Following Applications 92.1% IFEval score — best for agentic systems, tool use, and structured workflows. Benchmark

90-Day Price Trend

Input / Output price (USD per 1M tokens)

Past 90 days of records; every price change is shown here

Date Dimension Price (USD) Source
Output $0.400 OpenRouter
Input $0.130 OpenRouter
Output $0.400 OpenRouter
Input $0.130 OpenRouter
Output $0.400 OpenRouter
Input $0.130 OpenRouter
Output $0.400 OpenRouter
Input $0.130 OpenRouter
Output $0.400 OpenRouter
Input $0.130 OpenRouter
Output $0.400 OpenRouter
Input $0.130 OpenRouter
Output $0.400 OpenRouter
Input $0.130 OpenRouter
Output $0.400 OpenRouter
Input $0.130 OpenRouter
Output $0.400 OpenRouter
Input $0.130 OpenRouter
Output $0.400 OpenRouter
Input $0.130 OpenRouter
Output $0.400 OpenRouter
Input $0.130 OpenRouter
Output $0.320 OpenRouter
Input $0.100 OpenRouter
Output $0.320 OpenRouter
Input $0.100 OpenRouter
Output $0.320 OpenRouter
Input $0.100 OpenRouter
Output $0.320 OpenRouter
Input $0.100 OpenRouter
Output $0.320 OpenRouter
Input $0.100 OpenRouter
Output $0.320 OpenRouter
Input $0.100 OpenRouter
Output $0.320 OpenRouter
Input $0.100 OpenRouter
Output $0.320 OpenRouter
Input $0.100 OpenRouter
Output $0.320 OpenRouter
Input $0.100 OpenRouter
Output $0.320 OpenRouter
Input $0.100 OpenRouter
Output $0.320 OpenRouter
Input $0.100 OpenRouter
Output $0.320 OpenRouter
Input $0.100 OpenRouter
Output $0.320 OpenRouter
Input $0.100 OpenRouter
Output $0.320 OpenRouter
Input $0.100 OpenRouter

Key Insights

Key data points from this page for quick reference and citation.

  • Context window: 131,072 tokens
  • Provider: Meta
  • Model family: Llama 3.3
  • Modalities: Text
  • Data source: OpenRouter, updated daily