Version:

11B GPU: BF16

MAX Model

This version is not quantized and a GPU is recommended.

Install our magic package manager:
```
curl -ssL https://magic.modular.com/ | bash
```
Then run the source command that's printed in your terminal.

Install Max Pipelines in order to run this model.

magic global install max-pipelines && magic global update

Start a local endpoint for Llama-3.2-Vision-Instruct/11B:

max-pipelines serve --huggingface-repo-id=meta-llama/Llama-3.2-11B-Vision-Instruct

The endpoint is ready when you see the URI printed in your terminal:

Server ready on http://0.0.0.0:8000 (Press CTRL+C to quit)

Now open another terminal to send a request using curl:

curl -N http://0.0.0.0:8000/v1/chat/completions -H "Content-Type: application/json" -d '{
    "model": "meta-llama/Llama-3.2-11B-Vision-Instruct",
    "stream": true,
    "messages": [
        {
          "role": "user",
          "content": [
            {
              "type": "text",
              "text": "What is in this image?"
            },
            {
              "type": "image_url",
              "image_url": {
                "url": "http://images-assets.nasa.gov/image/iss072e571418/iss072e571418~orig.jpg"
              }
            }
          ]
        }
    ]

}' | grep -o '"content":"[^"]*"' | sed 's/"content":"//g' | sed 's/"//g' | tr -d '
' | sed 's/\n//g'

🎉 Hooray! You’re running Generative AI. Our goal is to make this as easy as possible.

Deploy this model to cloud

Llama 3.2-Vision Instruct

Llama 3.2-Vision Instruct, a sophisticated collection of multimodal large language models (LLMs), offers pretrained and instruction-tuned image reasoning with 11B and 90B configurations. These models excel in visual recognition, image reasoning, captioning, and answering general image-related queries, standing out in industry benchmarks.

Core Details:

Developer: Meta
Architecture: Built on Llama 3.1, featuring an auto-regressive language model using an optimized transformer architecture. A dedicated vision adapter integrates into the Llama 3.1 language model for image tasks.
Model Sizes: Available in 11B and 90B parameters.
Supported Languages: Primarily supports English for image+text tasks, with additional languages for text-only tasks.

Technical Specifications:

Model	Training Data	Params	Input Modalities	Output Modalities	Context Length	GQA	Data Volume	Knowledge Cutoff
Llama 3.2-Vision	Image, text pairs	11B	Text + Image	Text	128k	Yes	6B pairs	December 2023
Llama 3.2-Vision	Image, text pairs	90B	Text + Image	Text	128k	Yes	6B pairs	December 2023

Llama 3.2 is designed for versatility, supporting commercial and research use, specifically in visual recognition and reasoning tasks.

Intended Use

Intended Use Cases:

Llama 3.2-Vision serves both commercial and research sectors with a focus on visual tasks:

Visual Question Answering (VQA): Interprets images to answer related queries.
Document Visual Question Answering (DocVQA): Enhances understanding of text and context in documents.
Image Captioning: Narrates images by converting visuals into descriptive text.
Image-Text Retrieval: Facilitates matching between images and descriptive text.
Visual Grounding: Connects language with specific image regions for clarity and relevance.

Restrictions: The models are restricted from use beyond supported languages or in violation of laws and the Acceptable Use Policy.

Use and Implementation

The repository offers guidance for using Llama 3.2-Vision with compatible software tools, ensuring effective deployment in various applications.

Software Compatibility:

Ensure tools are updated for optimal performance with Llama 3.2-Vision.

Hardware and Software

Training Factors

Meta utilized proprietary GPU clusters and custom-built infrastructure for efficient pretraining and fine-tuning operations.

Component	Training Time (GPU hours)	Power Consumption (W)	Emissions (tons CO2eq)
Llama 3.2-vision 11B	Stage 1: 147K H100 hours	700	71
Llama 3.2-vision 90B	Stage 1: 885K H100 hours	700	513
Total	2.02M		584

Training energy usage and emissions are optimally managed within Meta's framework.

Training Data

Llama 3.2-Vision leverages 6 billion image and text pairs for pretraining, drawing from public datasets and synthetic examples for instruction tuning.

Benchmarks - Image Reasoning

Llama 3.2-Vision's performance on industry benchmarks underscores its strengths in visual comprehension tasks.

Base Pretrained Models

Category	Benchmark	Metric	Llama 3.2 11B	Llama 3.2 90B
Image Understanding	VQAv2 (val)	Accuracy	66.8	73.6
	Text VQA (val)	Relaxed accuracy	73.1	73.5
Visual Reasoning	MMMU (val, 0-shot)	Micro avg accuracy	41.7	49.3

Responsibility & Safety

Approach to Deployment

Llama 3.2 aligns safety strategies to ensure secure usage within diverse applications, minimizing risks through reinforced safety measures.

Responsible Deployment

Guidance and tools are provided to aid developers in implementing Llama for beneficial, secure applications with autonomy in safety management.

Community Engagement

Meta fosters an open, collaborative environment, promoting safe AI practices.

Partnerships and Grants:

Meta is part of industry groups for standardizing safety in AI practices and provides the Llama Impact Grants to support impactful AI applications.

Citations

View detailed documentation.

Metadata

architectures.0	MllamaForConditionalGeneration
model_type	mllama
processor_config.chat_template	{{- bos_token }} {%- if custom_tools is defined %} {%- set tools = custom_tools %} {%- endif %} {%- if not tools_in_user_message is defined %} {%- set tools_in_user_message = true %} {%- endif %} {%- if not date_string is defined %} {%- if strftime_now is defined %} {%- set date_string = strftime_now("%d %b %Y") %} {%- else %} {%- set date_string = "26 Jul 2024" %} {%- endif %} {%- endif %} {%- if not tools is defined %} {%- set tools = none %} {%- endif %} {#- This block extracts the system message, so we can slot it into the right place. #} {%- if messages[0]['role'] == 'system' %} {%- set system_message = messages[0]['content']\|trim %} {%- set messages = messages[1:] %} {%- set user_supplied_system_message = true %} {%- else %} {%- set system_message = "" %} {%- set user_supplied_system_message = false %} {%- endif %} {#- Find out if there are any images #} {% set image_ns = namespace(has_images=false) %} {%- for message in messages %} {%- for content in message['content'] %} {%- if content['type'] == 'image' %} {%- set image_ns.has_images = true %} {%- endif %} {%- endfor %} {%- endfor %} {#- System message if there are no images, or if the user supplied one #} {%- if user_supplied_system_message or not image_ns.has_images %} {{- "<\|start_header_id\|>system<\|end_header_id\|>\n\n" }} {%- if tools is not none %} {{- "Environment: ipython\n" }} {%- endif %} {{- "Cutting Knowledge Date: December 2023\n" }} {{- "Today Date: " + date_string + "\n\n" }} {%- if tools is not none and not tools_in_user_message %} {{- "You have access to the following functions. To call a function, please respond with JSON for a function call." }} {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }} {{- "Do not use variables.\n\n" }} {%- for t in tools %} {{- t \| tojson(indent=4) }} {{- "\n\n" }} {%- endfor %} {%- endif %} {{- system_message }} {{- "<\|eot_id\|>" }} {%- endif %} {#- Custom tools are passed in a user message with some extra guidance #} {%- if tools_in_user_message and not tools is none %} {#- Extract the first user message so we can plug it in here #} {%- if messages \| length != 0 %} {%- set first_user_message = messages[0]['content']\|trim %} {%- set messages = messages[1:] %} {%- else %} {{- raise_exception("Cannot put tools in the first user message when there's no first user message!") }} {%- endif %} {{- '<\|start_header_id\|>user<\|end_header_id\|>\n\n' -}} {{- "Given the following functions, please respond with a JSON for a function call " }} {{- "with its proper arguments that best answers the given prompt.\n\n" }} {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }} {{- "Do not use variables.\n\n" }} {%- for t in tools %} {{- t \| tojson(indent=4) }} {{- "\n\n" }} {%- endfor %} {{- first_user_message + "<\|eot_id\|>"}} {%- endif %} {%- for message in messages %} {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %} {{- '<\|start_header_id\|>' + message['role'] + '<\|end_header_id\|>\n\n' }} {%- if message['content'] is string %} {{- message['content'] }} {%- else %} {%- for content in message['content'] %} {%- if content['type'] == 'image' %} {{- '<\|image\|>' }} {%- elif content['type'] == 'text' %} {{- content['text'] }} {%- endif %} {%- endfor %} {%- endif %} {{- '<\|eot_id\|>' }} {%- elif 'tool_calls' in message %} {%- if not message.tool_calls\|length == 1 %} {{- raise_exception("This model only supports single tool-calls at once!") }} {%- endif %} {%- set tool_call = message.tool_calls[0].function %} {{- '<\|start_header_id\|>assistant<\|end_header_id\|>\n\n' -}} {{- '{"name": "' + tool_call.name + '", ' }} {{- '"parameters": ' }} {{- tool_call.arguments \| tojson }} {{- "}" }} {{- "<\|eot_id\|>" }} {%- elif message.role == "tool" or message.role == "ipython" %} {{- "<\|start_header_id\|>ipython<\|end_header_id\|>\n\n" }} {%- if message.content is mapping or message.content is iterable %} {{- message.content \| tojson }} {%- else %} {{- message.content }} {%- endif %} {{- "<\|eot_id\|>" }} {%- endif %} {%- endfor %} {%- if add_generation_prompt %} {{- '<\|start_header_id\|>assistant<\|end_header_id\|>\n\n' }} {%- endif %}

Version: 11B GPU BF16

This code works on compatible Linux machines.
We are actively working on enabling MAX Serve for MacOS ARM64 as well.

You can quickly deploy Llama-3.2-Vision-Instruct-11B to an endpoint using our MAX container. It includes the latest version of MAX with GPU support and our Python-based inference server called MAX Serve.

With the following Docker command, you’ll get an OpenAI-compatible endpoint running Llama-3.2-Vision-Instruct-11B:

docker run --gpus 1 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_HUB_ENABLE_HF_TRANSFER=1" \
    --env "HF_TOKEN=" \
    -p 8000:8000 \
    docker.modular.com/modular/max-openai-api:nightly \
    --huggingface-repo-id meta-llama/Llama-3.2-11B-Vision-Instruct

In order to download the model from Hugging Face, you just need to fill in the HF_TOKEN value with your access token, unless the model is from https://huggingface.co/modularai.

Learn more

For more information about the container image, see the MAX container documentation.

To learn more about how to deploy MAX to the cloud, check out our MAX Serve tutorials.

LLAMA 3.2 COMMUNITY LICENSE AGREEMENT Llama 3.2 Version Release Date: September 25, 2024

“Agreement” means the terms and conditions for use, reproduction, distribution and modification of the Llama Materials set forth herein.

“Documentation” means the specifications, manuals and documentation accompanying Llama 3.2 distributed by Meta at https://llama.meta.com/doc/overview.

“Licensee” or “you” means you, or your employer or any other person or entity (if you are entering into this Agreement on such person or entity’s behalf), of the age required under applicable laws, rules or regulations to provide legal consent and that has legal authority to bind your employer or such other person or entity if you are entering in this Agreement on their behalf.

“Llama 3.2” means the foundational large language models and software and algorithms, including machine-learning model code, trained model weights, inference-enabling code, training-enabling code, fine-tuning enabling code and other elements of the foregoing distributed by Meta at https://www.llama.com/llama-downloads.

“Llama Materials” means, collectively, Meta’s proprietary Llama 3.2 and Documentation (and any portion thereof) made available under this Agreement.

“Meta” or “we” means Meta Platforms Ireland Limited (if you are located in or, if you are an entity, your principal place of business is in the EEA or Switzerland) and Meta Platforms, Inc. (if you are located outside of the EEA or Switzerland).

By clicking “I Accept” below or by using or distributing any portion or element of the Llama Materials, you agree to be bound by this Agreement.

License Rights and Redistribution.

a. Grant of Rights. You are granted a non-exclusive, worldwide, non-transferable and royalty-free limited license under Meta’s intellectual property or other rights owned by Meta embodied in the Llama Materials to use, reproduce, distribute, copy, create derivative works of, and make modifications to the Llama Materials.

b. Redistribution and Use.
```
 i. If you distribute or make available the Llama Materials (or any derivative works thereof), 
```

or a product or service (including another AI model) that contains any of them, you shall (A) provide a copy of this Agreement with any such Llama Materials; and (B) prominently display “Built with Llama” on a related website, user interface, blogpost, about page, or product documentation. If you use the Llama Materials or any outputs or results of the Llama Materials to create, train, fine tune, or otherwise improve an AI model, which is distributed or made available, you shall also include “Llama” at the beginning of any such AI model name.

    ii. If you receive Llama Materials, or any derivative works thereof, from a Licensee as part

of an integrated end user product, then Section 2 of this Agreement will not apply to you.

    iii. You must retain in all copies of the Llama Materials that you distribute the

following attribution notice within a “Notice” text file distributed as a part of such copies: “Llama 3.2 is licensed under the Llama 3.2 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.”

    iv. Your use of the Llama Materials must comply with applicable laws and regulations

(including trade compliance laws and regulations) and adhere to the Acceptable Use Policy for the Llama Materials (available at https://www.llama.com/llama3_2/use-policy), which is hereby incorporated by reference into this Agreement.

Additional Commercial Terms. If, on the Llama 3.2 version release date, the monthly active users of the products or services made available by or for Licensee, or Licensee’s affiliates, is greater than 700 million monthly active users in the preceding calendar month, you must request a license from Meta, which Meta may grant to you in its sole discretion, and you are not authorized to exercise any of the rights under this Agreement unless or until Meta otherwise expressly grants you such rights.
Disclaimer of Warranty. UNLESS REQUIRED BY APPLICABLE LAW, THE LLAMA MATERIALS AND ANY OUTPUT AND RESULTS THEREFROM ARE PROVIDED ON AN “AS IS” BASIS, WITHOUT WARRANTIES OF ANY KIND, AND META DISCLAIMS ALL WARRANTIES OF ANY KIND, BOTH EXPRESS AND IMPLIED, INCLUDING, WITHOUT LIMITATION, ANY WARRANTIES OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE. YOU ARE SOLELY RESPONSIBLE FOR DETERMINING THE APPROPRIATENESS OF USING OR REDISTRIBUTING THE LLAMA MATERIALS AND ASSUME ANY RISKS ASSOCIATED WITH YOUR USE OF THE LLAMA MATERIALS AND ANY OUTPUT AND RESULTS.
Limitation of Liability. IN NO EVENT WILL META OR ITS AFFILIATES BE LIABLE UNDER ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, TORT, NEGLIGENCE, PRODUCTS LIABILITY, OR OTHERWISE, ARISING OUT OF THIS AGREEMENT, FOR ANY LOST PROFITS OR ANY INDIRECT, SPECIAL, CONSEQUENTIAL, INCIDENTAL, EXEMPLARY OR PUNITIVE DAMAGES, EVEN IF META OR ITS AFFILIATES HAVE BEEN ADVISED OF THE POSSIBILITY OF ANY OF THE FOREGOING.
Intellectual Property.

a. No trademark licenses are granted under this Agreement, and in connection with the Llama Materials, neither Meta nor Licensee may use any name or mark owned by or associated with the other or any of its affiliates, except as required for reasonable and customary use in describing and redistributing the Llama Materials or as set forth in this Section 5(a). Meta hereby grants you a license to use “Llama” (the “Mark”) solely as required to comply with the last sentence of Section 1.b.i. You will comply with Meta’s brand guidelines (currently accessible at https://about.meta.com/brand/resources/meta/company-brand/). All goodwill arising out of your use of the Mark will inure to the benefit of Meta.

b. Subject to Meta’s ownership of Llama Materials and derivatives made by or for Meta, with respect to any derivative works and modifications of the Llama Materials that are made by you, as between you and Meta, you are and will be the owner of such derivative works and modifications.

c. If you institute litigation or other proceedings against Meta or any entity (including a cross-claim or counterclaim in a lawsuit) alleging that the Llama Materials or Llama 3.2 outputs or results, or any portion of any of the foregoing, constitutes infringement of intellectual property or other rights owned or licensable by you, then any licenses granted to you under this Agreement shall terminate as of the date such litigation or claim is filed or instituted. You will indemnify and hold harmless Meta from and against any claim by any third party arising out of or related to your use or distribution of the Llama Materials.
Term and Termination. The term of this Agreement will commence upon your acceptance of this Agreement or access to the Llama Materials and will continue in full force and effect until terminated in accordance with the terms and conditions herein. Meta may terminate this Agreement if you are in breach of any term or condition of this Agreement. Upon termination of this Agreement, you shall delete and cease use of the Llama Materials. Sections 3, 4 and 7 shall survive the termination of this Agreement.
Governing Law and Jurisdiction. This Agreement will be governed and construed under the laws of the State of California without regard to choice of law principles, and the UN Convention on Contracts for the International Sale of Goods does not apply to this Agreement. The courts of California shall have exclusive jurisdiction of any dispute arising out of this Agreement.

DETAILS

VisionMODEL CLASS

MAX Model

MAX Models are popular open-source models converted to MAX’s native graph format. Anything with the label is either SOTA or being worked on. Learn more about MAX Models.

Browse all MAX Models

HARDWARE

GPU

QUANTIZATION

BF16

ARCHITECTURE

MAX Model

MAX GITHUB

Modular / MAX

MODEL

meta-llama

meta-llama/Llama-3.2-11B-Vision-Instruct

QUESTIONS ABOUT THIS MODEL?

Resources & support for
running Llama-3.2-Vision-Instruct-11B

Browse 27+ Tutorials

View Tutorials

Get help using MAX

Modular Forum

Read Documentation

Go to Docs

Llama-3.2-Vision-Instruct-11B

Llama 3.2-Vision Instruct

Core Details:

Technical Specifications:

Intended Use

Use and Implementation

Software Compatibility:

Hardware and Software

Training Factors

Training Data

Benchmarks - Image Reasoning

Base Pretrained Models

Responsibility & Safety

Approach to Deployment

Responsible Deployment

Community Engagement

Metadata

Learn more

Resources & support for running Llama-3.2-Vision-Instruct-11B

Resources & support for
running Llama-3.2-Vision-Instruct-11B