![](https://cdn-avatars.huggingface.co/v1/production/uploads/6435718aaaef013d1aec3b8b/XKf-8MA47tjVAM6SCX0MP.jpeg)
bartowski/Phi-3-mini-4k-instruct-GGUF
Text Generation
•
Updated
•
954
•
8
These quantized models have a smaller memory footprint, but acceptable quality.
Note Can run on devices with less memory. Let's say 4 GB RAM at q4 quantization, but you may be able to go as low as 2 GB RAM at q2 quantization.
Note Primary model of choice. Can run on PCs with 8 GB RAM at q4 quantization.
Note Honestly tied between this one and NeuralDaredevil. This one boasts a higher MMLU score which may confirm that it's a more knowledgeable model, but its prose is completely dry and it doesn't follow my requests as well.