Llama-3.2-90B-Vision-Instruct
Latest Llama-3.2-90B-Vision-InstructLarge open multimodal Llama model for image reasoning, document QA, and captioning
- Context window
- 128K tokens in · 4.1K out
The Llama-3.2-90B-Vision-Instruct line
2 releases in the Llama-3.2-90B-Vision-Instruct line.
BENCHMARKS
Not in the Epoch Capabilities Index. Epoch benchmarks what it can run, so this is a gap in coverage, not a poor result.
Meta's past year
8 release days in the past year, about one every 46 days. The next is expected around Dec 13, 2026. Forecasts ↗
NovDecJanFebMarAprMayJunJulAugSepOctNovDec
Specifications
RELEASE DATESep 25, 2024
WEIGHTSOpen weights
CONTEXT WINDOW128K
MAX OUTPUT4.1K
REASONINGNot listed
TOOL CALLINGNot listed
INPUTStext, image
OUTPUTStext
Specifications and prices from Models.dev; availability and limits may differ by provider. Benchmarks from Epoch AI (CC BY 4.0), LiveBench and the other sources named above.