How much does Llama 3.3 70B Instruct (free) cost?

Llama 3.3 70B Instruct (free) costs $0.00 per million input tokens and $0.00 per million output tokens. For a typical conversation (~2,000 tokens), that's approximately $0.000 per message.

What benchmarks has Llama 3.3 70B Instruct (free) been tested on?

Llama 3.3 70B Instruct (free) has been evaluated on 8 benchmarks. Top scores: MMLU: 81.7, MATH level 5: 41.6, Fiction.LiveBench: 33.3.

Is Llama 3.3 70B Instruct (free) open source?

Yes, Llama 3.3 70B Instruct (free) is open source.

How does Llama 3.3 70B Instruct (free) compare to Mistral 7B Instruct V0.2?

Llama 3.3 70B Instruct (free) has an average score of 29.6 while Mistral 7B Instruct V0.2 scores 29.7. Mistral 7B Instruct V0.2 slightly outperforms Llama 3.3 70B Instruct (free) overall. See full comparison →

Home/Models/Llama 3.3 70B Instruct (free)

Llama 3.3 70B Instruct (free)

Name: Llama 3.3 70B Instruct (free)
Author: Meta

by Meta · Released Dec 2024

Open Source

29.6

avg score

Rank #182

Compare

Better than 22% of all models

Context

66K tokens (~33 books)

Input $/1M

Free

Output $/1M

Free

Type

text

License

Open Source

Benchmarks

8 tested

Data updated today

About

The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out). The Llama 3.3 instruction tuned text only model...

Tested on 8 benchmarks with 29.1% average. Top scores: MMLU (81.7%), MATH level 5 (41.6%), Fiction.LiveBench (33.3%).

Capabilities

coding

14.4

#132 globally

reasoning

3.9

#165 globally

math

23.3

#150 globally

knowledge

42.0

#140 globally

Benchmark Scores

Compare All

Tested on 8 benchmarks · Ranked across 4 categories

Score Distribution (all 233 models)

0255075100

▲ You are here

codingCompare coding →

WeirdML

Unusual and adversarial machine learning challenges. Tests robustness of reasoning about edge cases in ML systems.

14.4—

reasoningCompare reasoning →

SimpleBench

Deceptively simple questions that humans find easy but AI models often get wrong. Tests common sense and reasoning gaps.

3.9—

mathCompare math →

MATH level 5

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

41.6—

OTIS Mock AIME 2024-2025

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

5.0—

Quick compare:

vs Mistral 7B Instruct V0.2

vs Claude 3 Haiku

vs Llama 3 70B Instruct

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

Model Family · Meta Llama 3.3

Llama 3.3 70B InstructDec 2024

46.9

$0.10/M in131Kctx8 benchmarks

Llama 3.3 70B Instruct (free)Dec 2024

29.1-17.8

$0.00/M in(-0.10)66Kctx(-66K)8 benchmarks

See the full Llama 3.3 family →

Similar Models

Mistral 7B Instruct V0.2

Frequently Asked Questions

Llama 3.3 70B Instruct (free) is an open-source text AI model by Meta, released in December 2024. It has an average benchmark score of 29.6. Context window: 66K tokens.

Benchmarks

MMLU MATH level 5 Fiction.LiveBench GPQA diamond Balrog

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

Llama 3.3 70B Instruct (free)

Frequently Asked Questions

Related Models

Benchmarks

Related Pages