Instructions to use NoesisLab/Kai-0.35B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use NoesisLab/Kai-0.35B-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="NoesisLab/Kai-0.35B-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("NoesisLab/Kai-0.35B-Instruct")
model = AutoModelForCausalLM.from_pretrained("NoesisLab/Kai-0.35B-Instruct", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use NoesisLab/Kai-0.35B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NoesisLab/Kai-0.35B-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/Kai-0.35B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/NoesisLab/Kai-0.35B-Instruct

SGLang

How to use NoesisLab/Kai-0.35B-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/Kai-0.35B-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/Kai-0.35B-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/Kai-0.35B-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/Kai-0.35B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use NoesisLab/Kai-0.35B-Instruct with Docker Model Runner:
```
docker model run hf.co/NoesisLab/Kai-0.35B-Instruct
```

OzTianlu commited on Feb 26

Commit

9fb36ef

verified ·

1 Parent(s): 41c199a

Update README.md

Browse files

Files changed (1) hide show

README.md +3 -3

README.md CHANGED Viewed

@@ -99,7 +99,7 @@ A compact 0.35B-parameter instruction-tuned language model optimized for reasoni
 2. **PIQA**: At **71.82%**, Kai-0.35B nearly matches TinyLlama-1.1B (73.0%) with only 1/3 the parameters, and trails the 1B-class Llama-3.2 by less than 3pp.
-3. **MBPP**: At **22.20%** pass@1, OpenKai-0.35B surpasses TinyLlama-1.1B (~19.91%) in code generation despite being 3x smaller.
 ## Usage
@@ -110,7 +110,7 @@ model = AutoModelForCausalLM.from_pretrained(
     "NoesisLab/Kai-0.35B-Instruct",
     torch_dtype=torch.bfloat16,
 )
-tokenizer = AutoTokenizer.from_pretrained("NoesisLab/OpenKai-0.35B-Instruct")
 messages = [{"role": "user", "content": "What is 25 * 4?"}]
 input_ids = tokenizer.apply_chat_template(messages, return_tensors="pt")
 output = model.generate(input_ids, max_new_tokens=256)
@@ -121,7 +121,7 @@ print(tokenizer.decode(output[0], skip_special_tokens=True))
 ## Citation
 ```bibtex
-@misc{noesislab2026openkai,
   title={Kai-0.35B-Instruct},
   author={NoesisLab},
   year={2026},

 2. **PIQA**: At **71.82%**, Kai-0.35B nearly matches TinyLlama-1.1B (73.0%) with only 1/3 the parameters, and trails the 1B-class Llama-3.2 by less than 3pp.
+3. **MBPP**: At **22.20%** pass@1, Kai-0.35B surpasses TinyLlama-1.1B (~19.91%) in code generation despite being 3x smaller.
 ## Usage
     "NoesisLab/Kai-0.35B-Instruct",
     torch_dtype=torch.bfloat16,
 )
+tokenizer = AutoTokenizer.from_pretrained("NoesisLab/Kai-0.35B-Instruct")
 messages = [{"role": "user", "content": "What is 25 * 4?"}]
 input_ids = tokenizer.apply_chat_template(messages, return_tensors="pt")
 output = model.generate(input_ids, max_new_tokens=256)
 ## Citation
 ```bibtex
+@misc{noesislab2026nkai,
   title={Kai-0.35B-Instruct},
   author={NoesisLab},
   year={2026},