Instructions to use concedo/OPT-19M-ChatSalad with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use concedo/OPT-19M-ChatSalad with Transformers:

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

pipe = pipeline("text-generation", model="concedo/OPT-19M-ChatSalad")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("concedo/OPT-19M-ChatSalad")
model = AutoModelForCausalLM.from_pretrained("concedo/OPT-19M-ChatSalad")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use concedo/OPT-19M-ChatSalad with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "concedo/OPT-19M-ChatSalad"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "concedo/OPT-19M-ChatSalad",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/concedo/OPT-19M-ChatSalad

SGLang

How to use concedo/OPT-19M-ChatSalad 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 "concedo/OPT-19M-ChatSalad" \
    --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": "concedo/OPT-19M-ChatSalad",
		"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 "concedo/OPT-19M-ChatSalad" \
        --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": "concedo/OPT-19M-ChatSalad",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use concedo/OPT-19M-ChatSalad with Docker Model Runner:
```
docker model run hf.co/concedo/OPT-19M-ChatSalad
```

OPT-19M-ChatSalad

This is an experimental OPT-based model with 19 million parameters trained entirely from scratch as a datasetting practice. Thus, it should not be subject to the usual OPT license. You are free to use this model for any purpose. The model is small enough (under 40mb) that it should run at very fast speeds even entirely on CPU.

It is recommend to use this model with the KoboldAI software, with the following parameters:

Temperature: 0.9
Repetition Penalty: 1.1
Repetition Penalty Slope: 0.7
Repetition Penalty Range: 1024
Top-P Sampling: 0.9
Disable all other samplers

All feedback and comments can be directed to Concedo on the KoboldAI discord.

Downloads last month: 1,298

concedo
/

OPT-19M-ChatSalad

OPT-19M-ChatSalad

Spaces using concedo/OPT-19M-ChatSalad 32