Meta: Llama 3.2 3B Instruct (free)

Meta: Llama 3.2 3B Instruct (free)

meta-llama · Released Sep 25, 2024 Emerging
Intelligence #502 / 583
15.9 Our Score
Speed #219 / 276
52.0 tokens / sec
Input
— Not priced
Output
— Not priced
Context #240 / 583
131,072 tokens

Analysis Summary

Meta's Llama 3.2 3B Instruct is one of the smallest models in the Llama 3.2 family, available free of charge. Its intelligence index of 4.2 and absence of a coding index reflect its limited capability for complex tasks, though its agentic index of 21.1 is surprisingly reasonable for its size.

For businesses, this model is best used as a fast, zero-cost classifier, intent router, or simple text formatter. It is not suitable for reasoning-heavy tasks, long-document analysis, or any coding workflow. Instruction-following scores are below average, so outputs require careful validation.

The free pricing makes it attractive for high-volume, low-stakes automation where accuracy requirements are modest. Teams should treat it as a utility model rather than a primary workhorse, pairing it with a stronger model for tasks requiring depth.

Assessed June 17, 2026

Editorial notes

Llama 3.2 3B Instruct is a free, ultra-lightweight Meta model suited to simple classification or routing tasks, with limited reasoning and no coding index data.

Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?

Performance Profile

Intelligence1.1Technical2.3Value0Content3.5
Intelligence 1.1/10
Technical 2.3/10
Content 3.5/10
Value 0/10

How Meta: Llama 3.2 3B Instruct (free) compares

Meta: Llama 3.2 3B Instruct (free) ranks #324 of 380 AI models we track for overall intelligence, #176 of 292 for agentic tasks. Its 131K-token context window is larger than 59% of the models we list. Meta: Llama 3.2 3B Instruct (free) is currently free to use via OpenRouter.

About Meta: Llama 3.2 3B Instruct (free)

Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization. Designed with the latest transformer architecture, it..

3B Parameters

Architecture Detail

Instruct Typellama3

Performance Indices

Source: Artificial Analysis

4.2 Intelligence Index
21.1 Agentic Index
3.3 Math Index

Benchmark Scores

Intelligence

GPQA Diamond 25.5% Graduate-level scientific reasoning
HLE 5.2% Humanity's Last Exam
MMLU Pro 34.7% Multi-task language understanding
MATH 500 48.9% Mathematical problem-solving
AIME 6.7% Competition mathematics
AIME 2025 3.3% Competition mathematics (2025)
SciCode 5.2% Scientific computing

Technical

LiveCodeBench 8.3% Live coding evaluation
τ²-Bench 21.1% Conversational agent benchmark

Content

IFBench 26.2% Instruction following
LCR 2% Long-context reasoning

Benchmark data from Artificial Analysis and Hugging Face

How does Meta: Llama 3.2 3B Instruct (free) stack up?

Compare side-by-side with other emerging models.

Compare Models

Model Information

OpenRouter ID meta-llama/llama-3.2-3b-instruct:free
Providermeta-llama
Model FamilyLlama 3
Release Date September 25, 2024
Context Length131,072 tokens
Status Active

Frequently asked questions about Meta: Llama 3.2 3B Instruct (free)

How much does Meta: Llama 3.2 3B Instruct (free) cost?

Meta: Llama 3.2 3B Instruct (free) is currently available for free via OpenRouter.

What is the context window of Meta: Llama 3.2 3B Instruct (free)?

Meta: Llama 3.2 3B Instruct (free) has a context window of 131,072 tokens (131K).

Who created Meta: Llama 3.2 3B Instruct (free)?

Meta: Llama 3.2 3B Instruct (free) is developed by Meta and was released on September 25, 2024.