Language model

Llama-3.2-11B-Vision-Instruct

Meta

Released: September 25, 2024
Data date: September 26, 2026

Llama 3.2 11B Vision Instruct adds image understanding to Meta’s smaller Llama line and processes text and images in one open checkpoint. That is what distinguishes it from Llama 3.2 3B, which is limited to text and code.

The weights use the Llama 3.2 Community License. The knowledge cutoff is December 2023.

Specifications and access

Specification	Value and source
Weights	Open weightsSource
Access	APISource
Parameters	11 billionSource
Context window	128,000 tokensSource
Maximum output	4,096 tokensSource
Knowledge Cutoff	December 2023Source
License	Llama 3.2 CommunitySource
Inputs	Text, ImageSource
Outputs	TextSource
Tool calling	YesSource

Sources and data date

Type	Evidence and data date
Additional source	models.dev catalog (retrieved September 26, 2026) · Meta model catalog snapshot · Editorial description reviewed October 2, 2026
Additional source	Hugging Face · Hugging Face, Llama 3.2 11B Vision Instruct · Editorial description reviewed October 2, 2026
Additional source	OpenRouter: Popularity index (derived), CC BY 4.0
Additional source	Vercel AI Gateway: Popularity index (derived), CC BY 4.0
Additional source	Creative Commons Attribution 4.0