Fazmin/solus_v1_deepseek-r1-distill-qwen-7b-q4

DeepSeek R1 Distill Qwen 7B — Solus v1

A reasoning model built by distilling DeepSeek-R1's chain-of-thought behaviour into a Qwen2.5 7B base. It works through problems step by step before committing to an answer, which makes it markedly stronger than similarly sized models on maths, logic puzzles, and multi-step analysis.

The trade-off is verbosity and latency: it spends tokens thinking, so it is slower to first useful output than a plain instruct model of the same size. MIT licensed.

Specifications

Single file: DeepSeek-R1-Distill-Qwen-7B-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: deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
• —Upstream GGUF: bartowski/DeepSeek-R1-Distill-Qwen-7B-GGUF
• —Mirrored for Solus, a desktop app for running language models entirely on your own machine.

Usage

llama-cli -m DeepSeek-R1-Distill-Qwen-7B-Q4_K_M.gguf -cnv