Skip to content
Local AI Wizard
Your machine
Theme

Llama-3.2-90B-Vision-Instruct

Meta · meta-llama/Llama-3.2-90B-Vision-Instruct

Weights license: Publisher terms: Llama 3.2 · code and components may use other terms

LMArena · vision1,032 ± 9#17 of 22 rated here · 2026-10-19
Memory needed~67 GBQ4 weights + working room, before the machine's reserve
Parameters88.6 BHub safetensors total
Repository updated2025-03-04A commit is not a release date

Quality

Reported by the publisher

12 results · Meta's model card · read 2026-10-25
MMMU (val, CoT)
60.3Image · College-level Problems and Mathematical Reasoning
MMMU-Pro, Standard (10 opts, test)
45.2Image · College-level Problems and Mathematical Reasoning
MMMU-Pro, Vision (test)
33.8Image · College-level Problems and Mathematical Reasoning
MathVista (testmini)
57.3Image · College-level Problems and Mathematical Reasoning
ChartQA (test, CoT)
85.5Image · Charts and Diagram Understanding
AI2 Diagram (test)
92.3Image · Charts and Diagram Understanding
DocVQA (test)
90.1Image · Charts and Diagram Understanding
VQAv2 (test)
78.1Image · General Visual Question Answering
Every reported result12
MMLU (CoT)
86.0Text · General
MATH (CoT)
68.0Text · Math
GPQA
46.7Text · Reasoning
MGSM (CoT)
86.9Text · Multilingual

Values as printed on the card, with the publisher's own settings (card section: Benchmarks - Image Reasoning › Instruction Tuned Models); column “Llama 3.2 90B”. Not comparable across publishers and not used in any ranking. Model card

Quality × memory

No comparable quality × memory pair

This entry has no score on a benchmark that is drawn against planning memory.

Consider it for

Meta released Llama-3.2-90B-Vision-Instruct on 2024-09-25: a vision-language model that answers about images, with 88.6B parameters on the Hugging Face count and the Publisher terms: Llama 3.2 license. Languages the publisher declares: en, de, fr, it, pt, hi, es, th.

Check first

License: commercial use allowed with conditions: Over 700M monthly users needs a Meta license; show "Built with Llama"; multimodal versions are not licensed to people or companies based in the EU (Acceptable Use Policy). At Q4 the weights load about 66.5 GB before each machine's reserve; image resolution, video frames and context add memory, and the vision encoder needs support in the runtime you pick. Downloading requires accepting the publisher's terms on Hugging Face.

Runs on

Fast memory holds it 15

The planning estimate fits the memory the model can use quickly: graphics card VRAM, or unified memory within the macOS GPU share.

Show all 15

Within memory, outside the fast part 1

Within the machine's memory, outside its fast part: on a graphics card the layers that do not fit run on the CPU from PC RAM (the listed RAM, or 32 GB assumed); on a Mac, above the macOS default GPU share. Not a speed figure.

One memory tier short 7

The same machines with the next memory size up would hold it in fast memory.

Show all 7

Hardware profiles grouped by the Q4 planning estimate on each machine (dedicated GPUs reserve 2 GB; unified and system memory 25%, 2 to 8 GB): the fast memory holds it, it is within memory but outside the fast part (layers offloaded to PC RAM, or above the macOS GPU share), or the machine is one memory tier short. Not a measured fit or speed.

Measured speed

No published measurement for this checkpoint

Tokens per second are shown only when a source publishes them with hardware, build and setup. They are never estimated.

How to run

Runtime compatibility and tokens per second have not been measured for this entry. Published metadata and a memory estimate do not prove a successful local run.

Library in source: transformers

Documented ways to run it

The model card could not be read (missing or gated)

Choose my computer

Technical details

What the source and the catalog store for this checkpoint. Missing fields are left out, never zero.

Task
image-text-to-text
Architecture
MllamaForConditionalGeneration
Library in source
transformers
Weight license
Publisher terms: Llama 3.2Llama 3.2 Vision model card (meta-llama/llama-models, GitHub) · read 2026-10-18source
Commercial use
Allowed with conditions Over 700M monthly users needs a Meta license; show "Built with Llama"; multimodal versions are not licensed to people or companies based in the EU (Acceptable Use Policy) · Llama 3.2 Community License (meta-llama/llama-models) · read 2026-10-18license text
Languages
en, de, fr, it, pt, hi, es, thModel card metadata · read 2026-10-25source
Released
Llama 3.2 Vision model card, Model Release Date (Meta, GitHub) · read 2026-10-18source
Every stored field11 fields
Revision checked
e305d2a43a4adc6987308fe7d896fb8ec5f1a5d8
Architecture
MllamaForConditionalGeneration
Library
transformers
Hub task tag
image-text-to-text
Formats
transformers, safetensors, pytorch
Access
Gated (manual)
Publisher kind
Official
Repository created
2024-09-19creation is not the model's release
Repository updated
2025-03-04a commit is not a release date
Hub downloads (30 days) · likes
499 · 366this repository only · read 2026-10-25
Catalog state
Hub config or weights found; no inventory benchmark link

Sources

Identity and published specifications come from the checkpoint page. Benchmark scores belong to the named release and do not measure speed on your computer.

Curated facts checked: 2026-10-18. The license and release date cite the publisher's sources; parameters, files and task come from the Hugging Face record, and memory is a labelled estimate from its parameter count. A value no source states stays empty; a repository commit is not a model release date.