Meta AI

Llama 3.1 70B

Type
language model
Context
131K tokens
Max output
128K
Released
23 July 2024
Knowledge cutoff
December 2023
API string
meta-llama/Llama-3.1-70B-Instruct
Weights
open

Prices by provider

Provider Input, $ per 1M Output, $ per 1M In our data since
Hyperbolic $0.12 $0.3 1 Aug 2026
Nebius Token Factory $0.13 $0.4 1 Aug 2026
DeepInfra $0.4 $0.4 1 Aug 2026
Kilo Code $0.4 $0.4 31 Jul 2026
OpenRouter $0.4 $0.4 31 Jul 2026
friendliai $0.6 $0.6 1 Aug 2026
OVHCloud AI Endpoints $0.67 $0.67 1 Aug 2026
Amazon Bedrock meta.llama3-1-70b-instruct-v1:0 $0.72 $0.72 2 Aug 2026
LLM Gateway $0.72 $0.72 31 Jul 2026
OCI Generative AI $0.72 $0.72 31 Jul 2026
Weights & Biases $0.8 $0.8 31 Jul 2026
Amazon Bedrock us.meta.llama3-1-70b-instruct-v1:0 $0.99 $0.99 2 Aug 2026
Perplexity $1 $1 31 Jul 2026
Microsoft Foundry $2.68 $3.54 1 Aug 2026

Only the base tier and only per-token prices are shown. Batch, discounted and cached rates, as well as prices per image or per second of video, do not go into this table: they cannot stand in the same column as a price per million tokens.

A single provider appears several times if it sells this model under different identifiers and at different prices — for example, at different weight precision. The identifier is shown next to the name. Identical offers that arrived from two sources under different spellings are merged into one row.

The date in the last column is the day this price first entered our collection. The price may well be older: before that day we simply were not recording it. It has not changed since — otherwise a new row with a new date would stand in its place.

Measurement results

Task set Result Run conditions Measured by
Arena Score in English 1,293.8 1,289–1,299
Arena Score, overall 1,261.07 1,257–1,265
Arena Score, programming 1,260.16 1,253–1,267
Arena Score in French 1,259.97 1,231–1,289
Arena Score, multi-turn dialogue 1,256.2 1,249–1,263
Arena Score in Spanish 1,252.23 1,227–1,277
Arena Score, mathematics 1,251.7 1,244–1,259
Arena Score, hard prompts 1,241.52 1,236–1,247
Arena Score, long queries 1,241.18 1,234–1,249
Arena Score in Russian 1,232.74 1,225–1,241
Arena Score, creative writing 1,232.27 1,225–1,240
Arena Score, instruction following 1,231.69 1,226–1,237
Arena Score, expert questions 1,208.23 1,197–1,220
MATH, difficulty level five 36.68 % · with a tuned harness Epoch evaluations
BALROG — game environments 27.9 % Balrog Leaderboard
GPQA Diamond — graduate-level questions 25.59 % · with a tuned harness Epoch evaluations
WeirdML — unusual machine learning tasks 8.97 % WeirdML Leaderboard
The Agent Company — work tasks in an office environment 6.9 % TheAgentCompany experiment results github
Mock AIME 2024–2025 — olympiad problems 3.52 % · with a tuned harness Epoch evaluations