How much does Llama 3 8B Instruct cost?

Llama 3 8B Instruct costs $0.03 per million input tokens and $0.04 per million output tokens. For a typical conversation (~2,000 tokens), that's approximately $0.000 per message.

What benchmarks has Llama 3 8B Instruct been tested on?

Llama 3 8B Instruct has been evaluated on 16 benchmarks. Top scores: Chatbot Arena Elo — Overall: 1222.2, ARC AI2: 77.1, OpenBookQA: 76.8.

Is Llama 3 8B Instruct open source?

Yes, Llama 3 8B Instruct is open source.

How does Llama 3 8B Instruct compare to Claude 2?

Llama 3 8B Instruct has an average score of 41.7 while Claude 2 scores 41.8. Claude 2 slightly outperforms Llama 3 8B Instruct overall. See full comparison →

Home/Models/Llama 3 8B Instruct

Llama 3 8B Instruct

Name: Llama 3 8B Instruct
Price: 0.03 USD
Author: Meta

by Meta · Released Apr 2024

Open Source

41.7

avg score

Rank #138

Compare

Better than 41% of all models

Context

8K tokens (~4 books)

Input $/1M

$0.03

Output $/1M

$0.04

Type

text

License

Open Source

Benchmarks

16 tested

Data updated today

About

Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors. This 8B instruct-tuned version was optimized for high quality dialogue usecases. It has demonstrated strong...

Tested on 16 benchmarks with 30.8% average. Top scores: Chatbot Arena Elo — Overall (1222.2%), ARC AI2 (77.1%), OpenBookQA (76.8%).

Capabilities

reasoning

19.9

#101 globally

math

3.6

#197 globally

knowledge

43.2

#133 globally

general

18.4

#43 globally

language

24.0

#132 globally

Benchmark Scores

Compare All

Tested on 16 benchmarks · Ranked across 6 categories

Score Distribution (all 233 models)

0255075100

▲ You are here

reasoningCompare reasoning →

MUSR

HuggingFace MuSR (Multi-Step Reasoning). Tests multi-hop reasoning requiring chaining multiple facts together.

19.9—

mathCompare math →

MATH level 5

Competition-level math from AMC, AIME, and olympiad problems. Level 5 is the hardest tier, requiring creative problem-solving.

6.1—

MATH Level 5

HuggingFace evaluation of MATH Level 5 problems. Competition math requiring advanced reasoning and proof construction.

3.9—

OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

0.7—

knowledgeCompare knowledge →

ARC AI2

AI2 Reasoning Challenge. Grade-school science questions requiring multi-step reasoning. Easy and Challenge sets test different difficulty levels.

77.1—

OpenBookQA

Elementary science questions with access to a small book of core science facts. Tests reasoning beyond memorization.

76.8—

TriviaQA

Trivia questions sourced from trivia enthusiasts and quiz websites. Tests breadth of general knowledge.

67.7—

Quick compare:

vs Claude 2

vs Gemini 1.5 Pro (Feb 2024)

vs GPT-4.1

Excellent (85+) Good (70-85) Average (50-70) Below (<50)

Model Family · Meta Llama 3

Llama 3 70B InstructApr 2024

32.4

$0.51/M in8Kctx9 benchmarks

Llama 3 8B InstructApr 2024

30.8-1.6

$0.03/M in(-0.48)8Kctx16 benchmarks

See the full Llama 3 family →

Similar Models

Claude 2

Anthropic

41.8TBD

Gemini 1.5 Pro (Feb 2024)

Links

Info

Meta Pricing explorer Developers · API

Research

Documentation

Community

Source Code

BenchGecko API

llama-3-8b-instruct

Specifications

Typetext
Context8K tokens (~4 books)
ReleasedApr 2024
LicenseOpen Source
StatusActive
Cost / Message~$0.000

Available On

Meta$0.03

Frequently Asked Questions

Llama 3 8B Instruct is an open-source text AI model by Meta, released in April 2024. It has an average benchmark score of 41.7. Context window: 8K tokens.

Benchmarks

Chatbot Arena Elo — Overall ARC AI2 OpenBookQA TriviaQA MMLU

Meta · Provider Meta · Economy All Models Compare Models Pricing Developers · API

Llama 3 8B Instruct

Frequently Asked Questions

Related Models

Benchmarks

Related Pages