Instructions to use Casual-Autopsy/CREC-n-WREC-Mate-24B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Casual-Autopsy/CREC-n-WREC-Mate-24B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Casual-Autopsy/CREC-n-WREC-Mate-24B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Casual-Autopsy/CREC-n-WREC-Mate-24B")
model = AutoModelForCausalLM.from_pretrained("Casual-Autopsy/CREC-n-WREC-Mate-24B", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Casual-Autopsy/CREC-n-WREC-Mate-24B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Casual-Autopsy/CREC-n-WREC-Mate-24B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Casual-Autopsy/CREC-n-WREC-Mate-24B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Casual-Autopsy/CREC-n-WREC-Mate-24B

SGLang

How to use Casual-Autopsy/CREC-n-WREC-Mate-24B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Casual-Autopsy/CREC-n-WREC-Mate-24B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Casual-Autopsy/CREC-n-WREC-Mate-24B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Casual-Autopsy/CREC-n-WREC-Mate-24B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Casual-Autopsy/CREC-n-WREC-Mate-24B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Casual-Autopsy/CREC-n-WREC-Mate-24B with Docker Model Runner:
```
docker model run hf.co/Casual-Autopsy/CREC-n-WREC-Mate-24B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

CREC-n-WREC-Mate-24B

This is a merge of pre-trained language models created using mergekit.

Merge Details

THIS MODEL IS UNOFFICIAL!

This model has no official affiliation with Weather and his SillyTavern Extensions. This is simply a fan project to help fellow users of these extensions.

Merge Description

CREC-n-WREC-Mate is a model made to help create World Info entries mid-roleplay using the SillyTavern extensions CREC and WREC.

The responses a bit on the shorter side by default, but this should be all the more beneficial for creating World Info entries. Needless to say, this isn't a model designed for creating Char Cards, instead it's meant for saving characters you encounter on your adventures to a Lorebook, so make sure to enable the feature that allows adding characters to a WI entry in the CREC settings menu.

Merge Method

This model was merged using the Conflict-Aware N:M Sparsification merge method using TheDrummer/Cydonia-24B-v2 as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: TheDrummer/Cydonia-24B-v2
  - model: AlexBefest/CardProjector-24B-v3
    parameters:
      weight: 0.4
      n_val: 16
      m_val: 32
  - model: CharGen/CharGen-v3-beta-rl-55-s48
    parameters:
      weight: 0.2
      n_val: 12
      m_val: 32
merge_method: cabs
default_n_val: 8
default_m_val: 32
pruning_order:
  - AlexBefest/CardProjector-24B-v3
  - CharGen/CharGen-v3-beta-rl-55-s48
base_model: TheDrummer/Cydonia-24B-v2
dtype: bfloat16