Instructions to use NoesisLab/Geilim-1B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use NoesisLab/Geilim-1B-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="NoesisLab/Geilim-1B-Instruct", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("NoesisLab/Geilim-1B-Instruct", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use NoesisLab/Geilim-1B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NoesisLab/Geilim-1B-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NoesisLab/Geilim-1B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/NoesisLab/Geilim-1B-Instruct

SGLang

How to use NoesisLab/Geilim-1B-Instruct 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 "NoesisLab/Geilim-1B-Instruct" \
    --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": "NoesisLab/Geilim-1B-Instruct",
		"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 "NoesisLab/Geilim-1B-Instruct" \
        --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": "NoesisLab/Geilim-1B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use NoesisLab/Geilim-1B-Instruct with Docker Model Runner:
```
docker model run hf.co/NoesisLab/Geilim-1B-Instruct
```

OzTianlu commited on Jan 30

Commit

e98313b

verified ·

1 Parent(s): 937e44a

Upload handler.py

Browse files

Files changed (1) hide show

handler.py +15 -12

handler.py CHANGED Viewed

@@ -101,18 +101,21 @@ class EndpointHandler:
             if _is_messages(item):
                 # Chat template path exists in repo; tokenizer.apply_chat_template will use it if configured
-                result = self.tokenizer.apply_chat_template(
-                    item,
-                    return_tensors="pt",
-                    add_generation_prompt=True,
-                )
-                # Handle both single tensor and tuple/dict returns
-                if isinstance(result, dict):
-                    input_ids = result["input_ids"]
-                elif isinstance(result, tuple):
-                    input_ids = result[0]
-                else:
-                    input_ids = result
             else:
                 if not isinstance(item, str):
                     item = str(item)

             if _is_messages(item):
                 # Chat template path exists in repo; tokenizer.apply_chat_template will use it if configured
+                try:
+                    # Use tokenize=False to get the formatted string first
+                    prompt = self.tokenizer.apply_chat_template(
+                        item,
+                        tokenize=False,
+                        add_generation_prompt=True,
+                    )
+                    # Then tokenize it separately to avoid unpacking issues
+                    enc = self.tokenizer(prompt, return_tensors="pt")
+                    input_ids = enc["input_ids"]
+                except Exception:
+                    # Fallback: if chat template fails, use the last user message
+                    last_user_msg = next((m["content"] for m in reversed(item) if m.get("role") == "user"), "")
+                    enc = self.tokenizer(last_user_msg, return_tensors="pt")
+                    input_ids = enc["input_ids"]
             else:
                 if not isinstance(item, str):
                     item = str(item)