Instructions to use teknium/Replit-v1-CodeInstruct-3B-fp16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use teknium/Replit-v1-CodeInstruct-3B-fp16 with Transformers:

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

pipe = pipeline("text-generation", model="teknium/Replit-v1-CodeInstruct-3B-fp16", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("teknium/Replit-v1-CodeInstruct-3B-fp16", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("teknium/Replit-v1-CodeInstruct-3B-fp16", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use teknium/Replit-v1-CodeInstruct-3B-fp16 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "teknium/Replit-v1-CodeInstruct-3B-fp16"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "teknium/Replit-v1-CodeInstruct-3B-fp16",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/teknium/Replit-v1-CodeInstruct-3B-fp16

SGLang

How to use teknium/Replit-v1-CodeInstruct-3B-fp16 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 "teknium/Replit-v1-CodeInstruct-3B-fp16" \
    --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": "teknium/Replit-v1-CodeInstruct-3B-fp16",
		"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 "teknium/Replit-v1-CodeInstruct-3B-fp16" \
        --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": "teknium/Replit-v1-CodeInstruct-3B-fp16",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use teknium/Replit-v1-CodeInstruct-3B-fp16 with Docker Model Runner:
```
docker model run hf.co/teknium/Replit-v1-CodeInstruct-3B-fp16
```

madhavatreplit commited on May 28, 2023

Commit

9c6e7d5

1 Parent(s): e5a060f

Fixing path issues to fix loading of models with Auto* classes.

Browse files

Was seeing errors in loading the tokenizer and models. This was because of misconfigured paths in the config.json.

This should fix them.

Files changed (1) hide show

config.json +2 -3

config.json CHANGED Viewed

@@ -1,5 +1,4 @@
 {
-  "_name_or_path": "./Replit-CodeInstruct",
   "architectures": [
     "MPTForCausalLM"
   ],
@@ -16,8 +15,8 @@
     "softmax_scale": null
   },
   "auto_map": {
-    "AutoConfig": "replit/replit-code-v1-3b--configuration_mpt.MPTConfig",
-    "AutoModelForCausalLM": "replit/replit-code-v1-3b--modeling_mpt.MPTForCausalLM"
   },
   "d_model": 2560,
   "emb_pdrop": 0,

 {
   "architectures": [
     "MPTForCausalLM"
   ],
     "softmax_scale": null
   },
   "auto_map": {
+    "AutoConfig": "configuration_mpt.MPTConfig",
+    "AutoModelForCausalLM": "modeling_mpt.MPTForCausalLM"
   },
   "d_model": 2560,
   "emb_pdrop": 0,