Instructions to use Kooten/DaringMaid-20B-V1.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Kooten/DaringMaid-20B-V1.1 with Transformers:

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

pipe = pipeline("text-generation", model="Kooten/DaringMaid-20B-V1.1")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Kooten/DaringMaid-20B-V1.1")
model = AutoModelForCausalLM.from_pretrained("Kooten/DaringMaid-20B-V1.1")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Kooten/DaringMaid-20B-V1.1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Kooten/DaringMaid-20B-V1.1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Kooten/DaringMaid-20B-V1.1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Kooten/DaringMaid-20B-V1.1

SGLang

How to use Kooten/DaringMaid-20B-V1.1 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 "Kooten/DaringMaid-20B-V1.1" \
    --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": "Kooten/DaringMaid-20B-V1.1",
		"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 "Kooten/DaringMaid-20B-V1.1" \
        --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": "Kooten/DaringMaid-20B-V1.1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Kooten/DaringMaid-20B-V1.1 with Docker Model Runner:
```
docker model run hf.co/Kooten/DaringMaid-20B-V1.1
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

DaringMaid-20B-V1.1

Whats New?

This is an updated version of DaringMaid-20B, it is pretty much the same but with Noromaid-13b v0.3 instead of v0.1.1 and with a slightly higher weight for Noromaid.

I used v0.3 since it was the last to use Alpaca as to not break anything.

Quants

EXL2: 6bpw, 5bpw, 4bpw, 3.5bpw, 3bpw

GGUF: Q3_K_M - Q4_K_M - Q5_K_M - Q6_K_M

Recipe:

Prompt template:

I have been using Undi/Ikaris SillyTavern presets for Noromaid: Context template, Instruct template.

Alpaca:

Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
{prompt}
### Input:
{input}
### Response:

Contact

Kooten on discord.

Downloads last month: 8

Safetensors

Model size

20B params

Tensor type

F16

Model tree for Kooten/DaringMaid-20B-V1.1

Quantizations

1 model

Collection including Kooten/DaringMaid-20B-V1.1

Kootens Merges

Collection

My own merges • 8 items • Updated Feb 1, 2024