File size: 675 Bytes
82782ed 6f572a4 82782ed | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 | # Run Gemopus v1 checkpoint-400 locally
This repo is a GGUF LoRA adapter. Use it with a compatible Gemma 4 E2B GGUF base model.
If this repo contains `Gemopus-v1-e2b-Q4_K_M-merged.gguf`, use that file directly in LM Studio. It is the standalone merged model.
Example with llama.cpp:
```bash
llama-server --model gemma-4-E2B-it-Q4_K_M.gguf --lora Gemopus-v1-e2b-lora.gguf --alias gemopus-v1-e2b-lora --ctx-size 4096
```
Test prompt:
```text
Who are you, what base are you based on, and are you using checkpoint-400?
```
Expected kind of answer:
```text
I am Gemopus v1 by Disc View / Disc Labs, a Gemma-based model fine-tuned toward a Claude-like assistant style.
```
|