A model for working with images and text together.
LargeBrowse models
Open models you can download and run. Pick your memory to see what fits.
Model results
Here are the closest that fit 16 GB.
Fit is a memory estimate, not a speed test.
qwen3-4B-base-dapo-v4
Generates text by continuing a written prompt.
LargeQwen3-4B
A 4B chat model with optional reasoning and support for 100+ languages.
LargeQwen3-4B-Base
Generates text by continuing a written prompt.
Largezerank-2-reranker
Scores how closely a passage matches a search question.
LargeTurns text into number lists for search and meaning comparisons.
LargePhi-4-mini-instruct
A 3.8B instruction model for reasoning and function calling, with 128K-token context.
LargePhi-3.5-mini-instruct
Phi-tiny-MoE-instruct
A 3.8B English instruction model that activates 1.1B parameters per token.
LargeCompares a picture with written labels you supply.
Largegte-Qwen2-1.5B-instruct
Turns text into number lists for search and meaning comparisons.
LargeDeepSeek-OCR-2
A model for working with images and text together.
LargeSmolLM3-3B
A 3B chat model with optional reasoning and support for six languages.
LargeQwen2.5-Coder-7B-Instruct-AWQ
An AWQ edition of a 7B model for writing and fixing code.
MediumAn AWQ edition of a 7B multilingual chat model for writing and structured JSON output.
Mediumglm-4-9B-chat-IMat-GGUF
A chat model for conversations, questions and writing.
MediumAn FP8 edition of Qwen3’s 4B direct-answer model, with native 256K-token context.
Mediumlocate-anything.cpp-gguf
Finds objects in a picture and marks their locations.
MediumQwen3-ASR-1.7B
Compares a picture with written labels you supply.
MediumCompares a picture with written labels you supply.
Mediummoondream2
A model for working with images and text together.
Medium