Instructions to use robowaifudev/megatron-gpt2-345m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use robowaifudev/megatron-gpt2-345m with Transformers:

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

pipe = pipeline("text-generation", model="robowaifudev/megatron-gpt2-345m")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("robowaifudev/megatron-gpt2-345m")
model = AutoModelForCausalLM.from_pretrained("robowaifudev/megatron-gpt2-345m")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use robowaifudev/megatron-gpt2-345m with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "robowaifudev/megatron-gpt2-345m"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "robowaifudev/megatron-gpt2-345m",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/robowaifudev/megatron-gpt2-345m

SGLang

How to use robowaifudev/megatron-gpt2-345m 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 "robowaifudev/megatron-gpt2-345m" \
    --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": "robowaifudev/megatron-gpt2-345m",
		"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 "robowaifudev/megatron-gpt2-345m" \
        --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": "robowaifudev/megatron-gpt2-345m",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use robowaifudev/megatron-gpt2-345m with Docker Model Runner:
```
docker model run hf.co/robowaifudev/megatron-gpt2-345m
```

robowaifudev commited on Dec 21, 2022

Commit

ce7f130

1 Parent(s): f803395

Update README.md

Browse files

Files changed (1) hide show

README.md +10 -0

README.md CHANGED Viewed

@@ -23,6 +23,16 @@ model-index:
           - type: wikitext
             value: 19.31
             name: Perplexity
 ---
 <!---

           - type: wikitext
             value: 19.31
             name: Perplexity
+      - task:
+          type: text-generation
+          name: Text generation
+        dataset:
+          name: WikiText-2
+          type: wikitext
+        metrics:
+          - type: wikitext
+            value: 17.151
+            name: Perplexity
 ---
 <!---