Llama 3.2 3B Instruct

Name: Llama 3.2 3B Instruct
Author: Meta AI

by Meta AILlama 3.2

Open WeightModel weights are publicly available — can be downloaded and self-hosted

Released September 2024131K context≈ 98,304 words3.2B params

A compact, nimble model that punches above its weight for everyday language tasks — following instructions, drafting short-form text, and handling conversational exchanges. Its small footprint makes it practical for edge deployment or resource-constrained environments, though it noticeably struggles with complex multi-step reasoning or deep domain knowledge that larger models handle more comfortably.

Capabilities

Capability scores are AI-generated based on model documentation, benchmarks, and technical specifications. Learn more

Instruction Following

Strong

Long Context

Llama 3.2 3B Instruct

by Meta AILlama 3.2

Open WeightModel weights are publicly available — can be downloaded and self-hosted

Released September 2024131K context≈ 98,304 words3.2B params

A compact, nimble model that punches above its weight for everyday language tasks — following instructions, drafting short-form text, and handling conversational exchanges. Its small footprint makes it practical for edge deployment or resource-constrained environments, though it noticeably struggles with complex multi-step reasoning or deep domain knowledge that larger models handle more comfortably.

Capabilities

Capability scores are AI-generated based on model documentation, benchmarks, and technical specifications. Learn more

Instruction Following

Strong

Long Context

Benchmark Scores

Benchmark	Score	Type	Recorded
GPQA Diamond	3.8	accuracy	26d ago
BBH	24.1	accuracy	26d ago
IFEval	73.9	accuracy	26d ago
MMLU-Pro	24.4	accuracy	26d ago
MuSR	1.4	accuracy	26d ago
MATH	17.7	accuracy	26d ago

Glossary

Domain KnowledgeSpecialized expertise and facts about a particular field or subject area that an AI model has learned during training.Edge DeploymentRunning a model directly on local devices like phones, tablets, or IoT hardware rather than sending data to a remote server.Multi-Step ReasoningThe ability to break down complex problems into smaller steps and solve them sequentially, rather than jumping directly to an answer.ReasoningThe model's ability to work through multi-step logical problems and provide justified answers rather than just pattern-matching.Resource-ConstrainedHardware with limited memory, processing power, or battery life, requiring models to be optimized for efficiency.

Capabilities

Capabilities

Benchmark Scores

Use Case Fit

Glossary