ZMIME
Meta · Released Jul 23, 2024

Llama-3.1-70B-Instruct

Open Llama instruction model for multilingual chat, reasoning, and coding.

  • Open weights
  • Tool calling
Capability
124.7ECI · #146 of 158
Input
$0.72per 1M tokens
Output
$0.72per 1M tokens
Context
128K4K max output
Pricing

What it costs

Input
$0.72per million tokens
Output
$0.72per million tokens
Blended (3:1)
$0.72cheaper than 51% of priced models

Typical monthly bills

Monthly usageEstimated cost
Side project2M input + 0.5M output tokens$1.80
Team assistant25M input + 5M output tokens$21.60
Production app250M input + 50M output tokens$216.00

Median of 7 paid API providers on models.dev.

Capability

Independent benchmarks

124.7Epoch Capabilities Index
#146 of 158 scored models
68Median 147174

The shaded band is Epoch AI’s confidence range (121.6–127.6); the tick marks the median scored model.

  • GPQA DiamondGraduate-level science questions44.2%
  • OTIS Mock AIME 2024–2025Competition mathematics3.6%

Scores from Epoch AI, run independently of Meta.

Specifications

The details

Lab
Meta
Released
Jul 23, 2024
Knowledge cutoff
Dec 2023
Context window
128,000 tokens
Max output
4,096 tokens
Inputs
Text
Output
Text
Reasoning
No
Tool calling
Yes
Structured output
No
Availability
7 API providerslisted on models.dev
Same lab

More from Meta

Questions

About Llama-3.1-70B-Instruct

How much does Llama-3.1-70B-Instruct cost?

Llama-3.1-70B-Instruct costs $0.72 input / $0.72 output per million tokens (median across 7 API providers). At a 3:1 input-to-output mix that is $0.72 per million tokens, cheaper than 51% of the 469 priced models we track.

What is the context window of Llama-3.1-70B-Instruct?

Llama-3.1-70B-Instruct accepts up to 128,000 tokens per request and can write up to 4,096 tokens in one response.

How good is Llama-3.1-70B-Instruct?

Epoch AI gives Llama-3.1-70B-Instruct a Capabilities Index score of 124.7 (likely range 121.6–127.6), ranking it #146 of 158 models Epoch has scored. Epoch AI benchmark results: GPQA Diamond 44.2%, OTIS Mock AIME 2024–2025 3.6%.

Is Llama-3.1-70B-Instruct open source?

Yes. Meta publishes the weights, so it can be downloaded and self-hosted.

What inputs does Llama-3.1-70B-Instruct support?

Llama-3.1-70B-Instruct accepts text and replies in text. It is not a dedicated reasoning model, supports tool calling.

When was Llama-3.1-70B-Instruct released?

Meta released Llama-3.1-70B-Instruct on Jul 23, 2024. Its training data runs to Dec 2023.

What are the best alternatives to Llama-3.1-70B-Instruct?

The closest current models from other labs on capability, price and release date are WizardLM 2 8x22B (Microsoft, $0.493 / $0.493), Claude Haiku 3.5 (Anthropic, ECI 124.8, $0.80 / $4.00), GPT-4o mini (OpenAI, ECI 123.8, $0.15 / $0.60) and Gemma 2 27B (Google, ECI 118.6, $0.65 / $0.65).