Instructions to use Katisim/Kat-Gen1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Katisim/Kat-Gen1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Katisim/Kat-Gen1")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("Katisim/Kat-Gen1") model = AutoModelForCausalLM.from_pretrained("Katisim/Kat-Gen1") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Katisim/Kat-Gen1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Katisim/Kat-Gen1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Katisim/Kat-Gen1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Katisim/Kat-Gen1
- SGLang
How to use Katisim/Kat-Gen1 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 "Katisim/Kat-Gen1" \ --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": "Katisim/Kat-Gen1", "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 "Katisim/Kat-Gen1" \ --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": "Katisim/Kat-Gen1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use Katisim/Kat-Gen1 with Docker Model Runner:
docker model run hf.co/Katisim/Kat-Gen1
Create model_card.json
Browse files- model_card.json +49 -0
model_card.json
ADDED
|
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_name": "Kat-Gen1",
|
| 3 |
+
"model_id": "Katisim/Kat-Gen1",
|
| 4 |
+
"model_type": "causal-lm",
|
| 5 |
+
"language": ["en"],
|
| 6 |
+
"license": "apache-2.0",
|
| 7 |
+
"tags": [
|
| 8 |
+
"text-generation",
|
| 9 |
+
"transformers",
|
| 10 |
+
"pytorch",
|
| 11 |
+
"causal-lm"
|
| 12 |
+
],
|
| 13 |
+
"datasets": [],
|
| 14 |
+
"metrics": [],
|
| 15 |
+
"pipeline_tag": "text-generation",
|
| 16 |
+
"widget": [
|
| 17 |
+
{
|
| 18 |
+
"text": "Once upon a time"
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"text": "The future of artificial intelligence"
|
| 22 |
+
}
|
| 23 |
+
],
|
| 24 |
+
"model_description": "Kat-Gen1 is a causal language model designed for efficient text generation across various natural language processing tasks.",
|
| 25 |
+
"intended_use": {
|
| 26 |
+
"primary_uses": [
|
| 27 |
+
"Text generation",
|
| 28 |
+
"Creative writing",
|
| 29 |
+
"Conversational AI",
|
| 30 |
+
"Content drafting"
|
| 31 |
+
],
|
| 32 |
+
"out_of_scope": [
|
| 33 |
+
"Medical advice",
|
| 34 |
+
"Legal advice",
|
| 35 |
+
"Real-time factual queries"
|
| 36 |
+
]
|
| 37 |
+
},
|
| 38 |
+
"limitations": [
|
| 39 |
+
"May generate biased content",
|
| 40 |
+
"Performance depends on prompt quality",
|
| 41 |
+
"Limited context window",
|
| 42 |
+
"Not suitable for factual accuracy-critical applications"
|
| 43 |
+
],
|
| 44 |
+
"ethical_considerations": [
|
| 45 |
+
"Content filtering recommended for production",
|
| 46 |
+
"May reflect training data biases",
|
| 47 |
+
"User monitoring advised"
|
| 48 |
+
]
|
| 49 |
+
}
|