---
base_model: ToastyPigeon/Beeper-King-22B
tags:
- llama-cpp
- gguf-my-repo
---

# Triangle104/Beeper-King-22B-Q4_K_M-GGUF
This model was converted to GGUF format from [`ToastyPigeon/Beeper-King-22B`](https://huggingface.co/ToastyPigeon/Beeper-King-22B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/ToastyPigeon/Beeper-King-22B) for more details on the model.

---
Model details:
-
This is another slightly-adjusted version of MS-Meadowlark-22B.

The models used in this merge are:

    nbeerbower/Mistral-Small-Gutenberg-Doppel-22B
    crestf411/MS-sunfall-v0.7.0
    Alfitaria/mistral-small-fujin-qlora
    ToastyPigeon/mistral-small-springdragon-qlora
    concedo/Beepo-22B

Specifically, changing the instruct portion and the base to apply the QLoRAs to from Mistral-Small-Instruct-2409 to Beepo-22B made the model a little more wild in a fun way. It's still done a good job of adhering to prompts and character descriptions, because Beepo is very compliant model.

Instruct format is Mistral V2 & V3, but thanks to the inclusion of Beepo this model should also work with Alpaca. 

---
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)

```bash
brew install llama.cpp

```
Invoke the llama.cpp server or the CLI.

### CLI:
```bash
llama-cli --hf-repo Triangle104/Beeper-King-22B-Q4_K_M-GGUF --hf-file beeper-king-22b-q4_k_m.gguf -p "The meaning to life and the universe is"
```

### Server:
```bash
llama-server --hf-repo Triangle104/Beeper-King-22B-Q4_K_M-GGUF --hf-file beeper-king-22b-q4_k_m.gguf -c 2048
```

Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.

Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```

Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```

Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Triangle104/Beeper-King-22B-Q4_K_M-GGUF --hf-file beeper-king-22b-q4_k_m.gguf -p "The meaning to life and the universe is"
```
or 
```
./llama-server --hf-repo Triangle104/Beeper-King-22B-Q4_K_M-GGUF --hf-file beeper-king-22b-q4_k_m.gguf -c 2048
```