# Llama 3.2 90B Vision [Canonical model page](https://headlessthrones.com/models/llama-3-2-90b-vision) · [Model library](https://headlessthrones.com/models) Public catalog reference from Headless Thrones. Exact catalog ID: `llama-3-2-90b-vision`. Catalog source-check date: 2026-10-23. This brief contains the public catalog answers and any catalog guide, not private drafts, historical editorial revisions or independently measured rankings. The canonical HTML page may also display separately published editorial. Current catalog answers ## What to know about Llama 3.2 90B Vision AI-compiled from linked model documentation. This is a catalog reference, not an independent performance review. Catalog checked October 23, 2026. [Exact model](https://headlessthrones.com/models/llama-3-2-90b-vision#model-answer-identity) · [Access](https://headlessthrones.com/models/llama-3-2-90b-vision#model-answer-access) · [Token limits](https://headlessthrones.com/models/llama-3-2-90b-vision#model-answer-context) · [Dates & knowledge](https://headlessthrones.com/models/llama-3-2-90b-vision#model-answer-dates) · [Inputs & outputs](https://headlessthrones.com/models/llama-3-2-90b-vision#model-answer-modalities) · [License](https://headlessthrones.com/models/llama-3-2-90b-vision#model-answer-license) · [Model details](https://headlessthrones.com/models/llama-3-2-90b-vision#model-answer-research) · [Weight details](https://headlessthrones.com/models/llama-3-2-90b-vision#model-answer-repository) ### Which Llama 3.2 90B Vision does this page describe? Llama 3.2 90B Vision is a model record from **Meta** in the Llama 3.2 family. [Source](https://huggingface.co/meta-llama/Llama-3.2-90B-Vision) The recorded model identifier or checkpoint is `meta-llama/Llama-3.2-90B-Vision`. Use that exact identifier when checking documentation or a provider listing. [Source](https://huggingface.co/meta-llama/Llama-3.2-90B-Vision) ### How can you access Llama 3.2 90B Vision? The catalog documents **Open weights** access for Llama 3.2 90B Vision. [Source](https://huggingface.co/meta-llama/Llama-3.2-90B-Vision) As of the catalog check, this version was listed as available. [Source](https://huggingface.co/meta-llama/Llama-3.2-90B-Vision) Open weights means the model weights are distributed for use under their stated terms. It does not establish a hosted API, an open-source license, or a hardware requirement. ### What are Llama 3.2 90B Vision's context and output limits? The publisher reports the context window as **128k tokens**. [Source](https://huggingface.co/meta-llama/Llama-3.2-90B-Vision/resolve/38850089f6a42b7394680c4e4bbd4c40fb62404f/README.md) We retain the publisher's notation and qualifiers; no exact token count or hosted serving allowance is inferred from it. A separate maximum-output value is not established in this record. The context or input figure should not be used as an output allowance. Context, input and output limits describe different constraints. Do not add the figures together to infer a larger window. Check the endpoint's accounting for conversation history, system instructions, tools, media and generated tokens; a documented maximum is not a guarantee of answer quality or a measured usable capacity. ### What dates are documented for Llama 3.2 90B Vision? **Knowledge cutoff: December 2023.** [Source](https://huggingface.co/meta-llama/Llama-3.2-90B-Vision/resolve/38850089f6a42b7394680c4e4bbd4c40fb62404f/README.md) This labels the documented knowledge boundary; it is not a promise that every earlier fact is known or correct. The source labels this “| Llama 3.2-Vision | (Image, text) pairs | 90B (88.8) | Text \\+ Image | Text | 128k | Yes | 6B (image, text) pairs | December 2023 |.” **Training-data cutoff: December 2023.** [Source](https://huggingface.co/meta-llama/Llama-3.2-90B-Vision/resolve/38850089f6a42b7394680c4e4bbd4c40fb62404f/README.md) This describes the reported data boundary, not the day training finished. The source labels this “Data Freshness: The pretraining data has a cutoff of December 2023\\..” **Release date: September 25, 2024.** [Source](https://huggingface.co/meta-llama/Llama-3.2-90B-Vision/resolve/38850089f6a42b7394680c4e4bbd4c40fb62404f/README.md) This is a release milestone, not a knowledge or training cutoff. The source labels this “Model Release Date: Sept 25, 2024.” Retrieval or web tools can supply newer information at run time. That does not change the model's recorded training or knowledge dates, and this record does not establish that a particular endpoint includes those tools. ### What can Llama 3.2 90B Vision receive and generate? Documented inputs: **Text and Image**. [Source](https://huggingface.co/meta-llama/Llama-3.2-90B-Vision/resolve/38850089f6a42b7394680c4e4bbd4c40fb62404f/README.md) Documented outputs: **Text**. [Source](https://huggingface.co/meta-llama/Llama-3.2-90B-Vision/resolve/38850089f6a42b7394680c4e4bbd4c40fb62404f/README.md) Input support and output generation are separate capabilities. Reading an image does not establish image generation; a text output listing does not establish speech output. File sizes, formats and media limits depend on the documented implementation. ### What license is recorded for Llama 3.2 90B Vision? The recorded license label is **llama3.2**. [Source](https://huggingface.co/api/models/meta-llama/Llama-3.2-90B-Vision) Read the linked license and associated use terms for this exact version before redistribution, commercial deployment or fine-tuning. A label alone does not establish every permission, restriction or obligation, and it does not determine the terms of a separate hosted service. ### What distinguishes Llama 3.2 90B Vision? These details come from the linked publisher material for this model. Reported capabilities and training descriptions are publisher statements, not independent test results. **Exact size specification row**: Publisher nominal size: 90B (88.8B). The family was pretrained on six billion image-text pairs and uses grouped-query attention. [Source](https://huggingface.co/meta-llama/Llama-3.2-90B-Vision/resolve/38850089f6a42b7394680c4e4bbd4c40fb62404f/README.md) **Vision language restriction**: Image-plus-text applications support English only; text-only use supports eight listed languages. [Source](https://huggingface.co/meta-llama/Llama-3.2-90B-Vision/resolve/38850089f6a42b7394680c4e4bbd4c40fb62404f/README.md) ### What is documented about the Llama 3.2 90B Vision checkpoint? This evidence refers to `meta-llama/Llama-3.2-90B-Vision` at recorded revision `38850089f6a42b7394680c4e4bbd4c40fb62404f`, checked October 23, 2026. [Repository metadata](https://huggingface.co/api/models/meta-llama/Llama-3.2-90B-Vision) Safetensors metadata records **88,593,355,323 stored parameters**. This is a stored-weight count, not necessarily the number activated per token, a download size or a GPU-memory requirement. [Repository metadata](https://huggingface.co/api/models/meta-llama/Llama-3.2-90B-Vision) Recorded architecture class: `MllamaForConditionalGeneration`. [Repository metadata](https://huggingface.co/api/models/meta-llama/Llama-3.2-90B-Vision) Repository task tag: `image-text-to-text`. A task tag identifies a distribution category; it does not measure performance on that task. [Repository metadata](https://huggingface.co/api/models/meta-llama/Llama-3.2-90B-Vision) Listed library: `transformers`. Check that library's version and model support before loading the weights. [Repository metadata](https://huggingface.co/api/models/meta-llama/Llama-3.2-90B-Vision) Repository language labels: en, de, fr, it, pt, hi, es, and th. These labels are not independent evaluations of language quality. [Repository metadata](https://huggingface.co/api/models/meta-llama/Llama-3.2-90B-Vision) Repository access gate: `manual`. This records the repository's gate setting when checked; licensing and use terms still apply. [Repository metadata](https://huggingface.co/api/models/meta-llama/Llama-3.2-90B-Vision) Read the retrieved model card for setup and qualifications. [Model card](https://huggingface.co/meta-llama/Llama-3.2-90B-Vision/resolve/38850089f6a42b7394680c4e4bbd4c40fb62404f/README.md) **Catalog qualifications:** Official vision checkpoint; exact limits not copied from the separate Llama 3.2 text-only family card. [Source 1](https://huggingface.co/api/models/meta-llama/Llama-3.2-90B-Vision) · [Source 2](https://huggingface.co/meta-llama/Llama-3.2-90B-Vision/resolve/38850089f6a42b7394680c4e4bbd4c40fb62404f/README.md) **Evidence still missing here:** tested hardware requirements. Missing documentation is not evidence that a feature is unavailable. Headless Thrones has not independently benchmarked this model for this reference. Evaluate it on representative work before choosing it; these facts do not rank its accuracy, reliability, speed or suitability against another model. Keep the exact version in view ## Other Llama 3.2 models from Meta. These are separate records in the same publisher family. A shared family name does not establish equal capabilities or interchangeable licenses. ### [Llama 3.2 11B Vision](https://headlessthrones.com/models/llama-3-2-11b-vision) Open weights ### [Llama 3.2 11B Vision Instruct](https://headlessthrones.com/models/llama-3-2-11b-vision-instruct) Open weights ### [Llama 3.2 90B Vision Instruct](https://headlessthrones.com/models/llama-3-2-90b-vision-instruct) Open weights ### [Llama 3.2 1B](https://headlessthrones.com/models/llama-3-2-1b) Open weights · 128,000 context tokens ### [Llama 3.2 1B Instruct](https://headlessthrones.com/models/llama-3-2-1b-instruct) Open weights · 128,000 context tokens ### [Llama 3.2 3B](https://headlessthrones.com/models/llama-3-2-3b) Open weights · 128,000 context tokens