Instructions to use Azrail/smallm_70 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Azrail/smallm_70 with Transformers:

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

pipe = pipeline("text-generation", model="Azrail/smallm_70", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("Azrail/smallm_70", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Azrail/smallm_70 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Azrail/smallm_70"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Azrail/smallm_70",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Azrail/smallm_70

SGLang

How to use Azrail/smallm_70 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 "Azrail/smallm_70" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Azrail/smallm_70",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "Azrail/smallm_70" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Azrail/smallm_70",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Azrail/smallm_70 with Docker Model Runner:
```
docker model run hf.co/Azrail/smallm_70
```

Azrail commited on Apr 15, 2025

Commit

0a6c273

verified ·

1 Parent(s): 34f3cd9

Upload tokenizer

Browse files

Files changed (3) hide show

special_tokens_map.json +4 -0
tokenizer.json +4 -4
tokenizer_config.json +7 -2

special_tokens_map.json CHANGED Viewed

@@ -1,4 +1,8 @@
 {
   "bos_token": {
     "content": "<|beginoftext|>",
     "lstrip": false,

 {
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
   "bos_token": {
     "content": "<|beginoftext|>",
     "lstrip": false,

tokenizer.json CHANGED Viewed

@@ -23,7 +23,7 @@
     },
     {
       "id": 2,
-      "content": "<|reserved_token_1|>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -32,7 +32,7 @@
     },
     {
       "id": 3,
-      "content": "<|reserved_token_2|>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -215,8 +215,8 @@
     "vocab": {
       "<|endoftext|>": 0,
       "<|beginoftext|>": 1,
-      "<|reserved_token_1|>": 2,
-      "<|reserved_token_2|>": 3,
       "<|reserved_token_3|>": 4,
       "<|reserved_token_4|>": 5,
       "<|reserved_token_5|>": 6,

     },
     {
       "id": 2,
+      "content": "<|im_start|>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
     },
     {
       "id": 3,
+      "content": "<|im_end|>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
     "vocab": {
       "<|endoftext|>": 0,
       "<|beginoftext|>": 1,
+      "<|im_start|>": 2,
+      "<|im_end|>": 3,
       "<|reserved_token_3|>": 4,
       "<|reserved_token_4|>": 5,
       "<|reserved_token_5|>": 6,

tokenizer_config.json CHANGED Viewed

@@ -17,7 +17,7 @@
       "special": true
     },
     "2": {
-      "content": "<|reserved_token_1|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -25,7 +25,7 @@
       "special": true
     },
     "3": {
-      "content": "<|reserved_token_2|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -105,7 +105,12 @@
       "special": true
     }
   },
   "bos_token": "<|beginoftext|>",
   "clean_up_tokenization_spaces": false,
   "eos_token": "<|endoftext|>",
   "extra_special_tokens": {},

       "special": true
     },
     "2": {
+      "content": "<|im_start|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
       "special": true
     },
     "3": {
+      "content": "<|im_end|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
       "special": true
     }
   },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
   "bos_token": "<|beginoftext|>",
+  "chat_template": "{% for message in messages %}{% if message.get('role') is not none %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% else %}{{message['content'] + '<|im_end|>' + '\n'}}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
   "clean_up_tokenization_spaces": false,
   "eos_token": "<|endoftext|>",
   "extra_special_tokens": {},