Instructions to use DiscoResearch/DiscoLM-70b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DiscoResearch/DiscoLM-70b with Transformers:

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

pipe = pipeline("text-generation", model="DiscoResearch/DiscoLM-70b")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("DiscoResearch/DiscoLM-70b")
model = AutoModelForCausalLM.from_pretrained("DiscoResearch/DiscoLM-70b")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use DiscoResearch/DiscoLM-70b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DiscoResearch/DiscoLM-70b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DiscoResearch/DiscoLM-70b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/DiscoResearch/DiscoLM-70b

SGLang

How to use DiscoResearch/DiscoLM-70b 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 "DiscoResearch/DiscoLM-70b" \
    --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": "DiscoResearch/DiscoLM-70b",
		"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 "DiscoResearch/DiscoLM-70b" \
        --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": "DiscoResearch/DiscoLM-70b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use DiscoResearch/DiscoLM-70b with Docker Model Runner:
```
docker model run hf.co/DiscoResearch/DiscoLM-70b
```

Fix Stop Tokens

by CyberTimon - opened Dec 6, 2023

base: refs/heads/main

←

from: refs/pr/2

Discussion Files changed

-3

Files changed (2) hide show

config.json +2 -2
tokenizer_config.json +1 -1

config.json CHANGED Viewed

@@ -4,8 +4,8 @@
     "LlamaForCausalLM"
   ],
   "attention_bias": false,
-  "bos_token_id": 1,
-  "eos_token_id": 2,
   "hidden_act": "silu",
   "hidden_size": 8192,
   "initializer_range": 0.02,

     "LlamaForCausalLM"
   ],
   "attention_bias": false,
+  "bos_token_id": 32001,
+  "eos_token_id": 32000,
   "hidden_act": "silu",
   "hidden_size": 8192,
   "initializer_range": 0.02,

tokenizer_config.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
   "add_bos_token": true,
-  "add_eos_token": false,
   "added_tokens_decoder": {
     "0": {
       "content": "<unk>",

 {
   "add_bos_token": true,
+  "add_eos_token": true,
   "added_tokens_decoder": {
     "0": {
       "content": "<unk>",