Modelpublic

Fazmin/solus_v1_phi-3.5-mini-instruct-q4

sourceHugging Facemitupdated 2mo agoView on Hugging Face

0likes10downloads

Model Card

Phi 3.5 Mini Instruct — Solus v1

Microsoft's Phi-3.5-mini, a 3.8B parameter model trained largely on filtered public data and synthetic textbook-style material. That data recipe gives it reasoning and comprehension quality well above what its size suggests, and it is particularly good at reading a document and answering questions about it.

It is multilingual, permissively MIT licensed, and a strong pick for summarisation and document Q&A on modest hardware.

Specifications


Parameters	3.8B
Quantization	Q4KM
File size	2.23 GB
Minimum RAM	6.00 GB
Minimum VRAM	not required
Context length	16,384 tokens
SHA-256	`e4165e3a71af97f1b4820da61079826d8752a2088e313af0c7d346796c38eff5`

Single file: Phi-3.5-mini-instruct-Q4_K_M.gguf

Quantization

Quantization performed at the Faculty of Engineering, McMaster University.

The GGUF conversion this build is derived from was produced by bartowski, and the weights here are a byte-for-byte copy of that file — the SHA-256 above matches the upstream artifact.

Provenance

—Original model: microsoft/Phi-3.5-mini-instruct
—Upstream GGUF: bartowski/Phi-3.5-mini-instruct-GGUF
—Mirrored for Solus, a desktop app for running language models entirely on your own machine.

Usage

bash

llama-cli -m Phi-3.5-mini-instruct-Q4_K_M.gguf -cnv