Instructions to use KoboldAI/GPT-Neo-125M-AID with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use KoboldAI/GPT-Neo-125M-AID with Transformers:

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

pipe = pipeline("text-generation", model="KoboldAI/GPT-Neo-125M-AID")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("KoboldAI/GPT-Neo-125M-AID")
model = AutoModelForCausalLM.from_pretrained("KoboldAI/GPT-Neo-125M-AID")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use KoboldAI/GPT-Neo-125M-AID with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KoboldAI/GPT-Neo-125M-AID"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KoboldAI/GPT-Neo-125M-AID",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/KoboldAI/GPT-Neo-125M-AID

SGLang

How to use KoboldAI/GPT-Neo-125M-AID 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 "KoboldAI/GPT-Neo-125M-AID" \
    --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": "KoboldAI/GPT-Neo-125M-AID",
		"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 "KoboldAI/GPT-Neo-125M-AID" \
        --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": "KoboldAI/GPT-Neo-125M-AID",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use KoboldAI/GPT-Neo-125M-AID with Docker Model Runner:
```
docker model run hf.co/KoboldAI/GPT-Neo-125M-AID
```

Henk717 commited on Dec 24, 2021

Commit

409e074

1 Parent(s): 264a571

Add KoboldAI values

Browse files

This defaults repetition penalty to 2.0 and enables Adventure mode by default if used with KoboldAI

Files changed (1) hide show

config.json +3 -1

config.json CHANGED Viewed

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "EleutherAI/gpt-neo-125M",
   "activation_function": "gelu_new",
   "architectures": [
     "GPTNeoForCausalLM"
@@ -49,5 +49,7 @@
   "transformers_version": "4.6.1",
   "use_cache": true,
   "vocab_size": 50257,
   "window_size": 256
 }

 {
+  "_name_or_path": "KoboldAI/GPT-Neo-125M-AID",
   "activation_function": "gelu_new",
   "architectures": [
     "GPTNeoForCausalLM"
   "transformers_version": "4.6.1",
   "use_cache": true,
   "vocab_size": 50257,
+  "rep_pen": 2.0,
+  "adventure": true,
   "window_size": 256
 }