Instructions to use parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft with Transformers:

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

pipe = pipeline("text-generation", model="parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft", trust_remote_code=True)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft

SGLang

How to use parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft 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 "parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft" \
    --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": "parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft",
		"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 "parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft" \
        --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": "parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft with Docker Model Runner:
```
docker model run hf.co/parallelstudios/mpt-7b-instruct-parallel-colony-memory-importance-ft
```

Schmadge commited on Jul 7, 2023

Commit

cd95ea8

1 Parent(s): 9e81bca

Update handler.py

Browse files

Files changed (1) hide show

handler.py +4 -10

handler.py CHANGED Viewed

@@ -4,21 +4,15 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 from typing import Any, Dict
 class EndpointHandler:
-    def __init__(
-        self,
-        path='',
-        torch_dtype=torch.bfloat16,
-        trust_remote_code=True,
-    ) -> None:
         self.model = AutoModelForCausalLM.from_pretrained(
             path,
             torch_dtype=torch_dtype,
             trust_remote_code=trust_remote_code
         )
-        tokenizer = AutoTokenizer.from_pretrained(
-            path,
-            trust_remote_code=trust_remote_code
-        )
         if tokenizer.pad_token_id is None:
             warnings.warn(
                 "pad_token_id is not set for the tokenizer. Using eos_token_id as pad_token_id."

 from typing import Any, Dict
 class EndpointHandler:
+    def __init__(self, path='', torch_dtype=torch.bfloat16, trust_remote_code=True):
         self.model = AutoModelForCausalLM.from_pretrained(
             path,
             torch_dtype=torch_dtype,
             trust_remote_code=trust_remote_code
         )
+        tokenizer = AutoTokenizer.from_pretrained("EleutherAI/gpt-neox-20b")
         if tokenizer.pad_token_id is None:
             warnings.warn(
                 "pad_token_id is not set for the tokenizer. Using eos_token_id as pad_token_id."