Instructions to use kd13/Type-o1-nano-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kd13/Type-o1-nano-instruct with Transformers:

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

pipe = pipeline("text-generation", model="kd13/Type-o1-nano-instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kd13/Type-o1-nano-instruct")
model = AutoModelForCausalLM.from_pretrained("kd13/Type-o1-nano-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 kd13/Type-o1-nano-instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kd13/Type-o1-nano-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": "kd13/Type-o1-nano-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/kd13/Type-o1-nano-instruct

SGLang

How to use kd13/Type-o1-nano-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 "kd13/Type-o1-nano-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": "kd13/Type-o1-nano-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 "kd13/Type-o1-nano-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": "kd13/Type-o1-nano-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use kd13/Type-o1-nano-instruct with Docker Model Runner:
```
docker model run hf.co/kd13/Type-o1-nano-instruct
```

kd13 commited on Jul 12

Commit

cf9ecc0

verified ·

1 Parent(s): 928eb9f

Create README.md

Browse files

Files changed (1) hide show

README.md +113 -0

README.md ADDED Viewed

	@@ -0,0 +1,113 @@

+---
+license: mit
+language:
+- en
+- hi
+base_model:
+- Qwen/Qwen2.5-0.5B
+pipeline_tag: text-generation
+library_name: transformers
+tags:
+- qwen
+- nano
+---
+# Type-o1-nano-instruct
+A very compact general-purpose instruct model designed for lightweight assistant use across everyday tasks — general chat, Python coding help, writing and content generation, language tasks, math, and tool-style web search workflows.
+The model is intended for small-footprint deployments where users need clear, practical answers and helpful explanations without the cost of a larger model.
+## Capabilities
+This model can help with:
+* General chat and multi-turn conversation
+* Python coding assistance and code explanation
+* Mathematics and basic quantitative reasoning
+* Engineering concepts and explanations
+* Creative writing (stories, poetry, writing prompts)
+* Content generation (marketing copy, social media captions, emails)
+* English grammar correction and rewriting
+* Advanced NLP tasks:
+  * Fill-mask
+  * Table question answering
+  * Context-based question answering (SQuAD style)
+  * Summarization (dialogue, news, and scientific papers)
+* English ↔ Hindi translation
+* Web search tool-call style conversations
+## Chat Format
+The model follows a Harmony-style chat structure.
+Supported interaction flow:
+```text
+system -> developer -> user -> tool call -> tool result -> final response
+```
+For normal chat use, you can use a standard chat-template style prompt.
+## Web Search Tool-Call Style
+The model can be used in tool-calling style conversations where the assistant decides when a search is needed, emits a tool call, receives a tool result, and then writes the final answer.
+Example structure:
+```text
+system: You are a helpful assistant with access to web search.
+user: Find the latest information about a topic.
+assistant tool call: web_search(...)
+tool result: ...
+assistant final: Answer using the search result.
+```
+Actual tool execution depends on your inference framework or application wrapper.
+## Recommended Use Cases
+This model is best suited for:
+* Lightweight general-purpose assistants
+* On-device or low-resource deployments
+* Writing and content generation helpers
+* Grammar and language correction tools
+* English ↔ Hindi translation helpers
+* Summarization and document Q&A tools
+* Beginner Python learning assistants
+* Tool-call research experiments
+* Chatbots where speed and small size matter more than depth
+## Limitations
+This model is not recommended for:
+* Production-critical software generation without review
+* Non-Python coding tasks such as C++, Java, Rust, Go, or JavaScript
+* Security-sensitive code generation
+* Medical, legal, or financial decision-making
+* Advanced or research-level mathematics
+* Long multi-file software engineering tasks
+* Tasks requiring very long context
+* High-stakes factual lookup without verification
+The model may sometimes:
+* Produce incorrect facts or reasoning
+* Miss edge cases
+* Over-explain simple questions
+* Generate code that needs testing
+* Struggle with very long context
+* Use tool-call format inconsistently depending on the prompt
+* Give uneven quality across its supported domains
+Being a very small model, it is best used for straightforward tasks rather than complex or nuanced ones. Always verify important outputs and test generated code before using it.
+## License
+Please check the model repository license before commercial or production use.
+## Disclaimer
+This model is an experimental small general-purpose assistant. It should be used as a helpful assistant, not as a guaranteed source of truth. For important tasks, verify outputs with tests, documentation, and human review.