Text Generation
Transformers
Safetensors
Portuguese
nanothink
thinking
reasoning
reason
think
lowparams
5m_params
thinkset-ptbr
gpt2
Instructions to use AxionLab-Co/NanoThink-5M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AxionLab-Co/NanoThink-5M with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="AxionLab-Co/NanoThink-5M")# Load model directly from transformers import NanoThink model = NanoThink.from_pretrained("AxionLab-Co/NanoThink-5M", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use AxionLab-Co/NanoThink-5M with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "AxionLab-Co/NanoThink-5M" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "AxionLab-Co/NanoThink-5M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/AxionLab-Co/NanoThink-5M
- SGLang
How to use AxionLab-Co/NanoThink-5M 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 "AxionLab-Co/NanoThink-5M" \ --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": "AxionLab-Co/NanoThink-5M", "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 "AxionLab-Co/NanoThink-5M" \ --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": "AxionLab-Co/NanoThink-5M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use AxionLab-Co/NanoThink-5M with Docker Model Runner:
docker model run hf.co/AxionLab-Co/NanoThink-5M
Update README.md
Browse files
README.md
CHANGED
|
@@ -96,7 +96,7 @@ João tem 3 maçãs e ganhou 2, quantas ele tem agora?
|
|
| 96 |
3 + 2 = 5
|
| 97 |
</THINK>
|
| 98 |
<ANSWER>
|
| 99 |
-
João
|
| 100 |
</ANSWER>
|
| 101 |
```
|
| 102 |
|
|
@@ -147,15 +147,16 @@ model.eval()
|
|
| 147 |
## 🔮 Future Work
|
| 148 |
|
| 149 |
* Scaling to 10M–50M parameters
|
| 150 |
-
* Improving dataset quality
|
| 151 |
* Enhancing reasoning consistency
|
| 152 |
* Multilingual support
|
| 153 |
|
|
|
|
| 154 |
---
|
| 155 |
|
| 156 |
## 🤝 Contributions
|
| 157 |
|
| 158 |
-
This is an experimental project
|
| 159 |
|
| 160 |
---
|
| 161 |
|
|
@@ -168,6 +169,7 @@ MIT
|
|
| 168 |
## 🧠 Author
|
| 169 |
|
| 170 |
AxionLab Co.
|
|
|
|
| 171 |
Independent research project exploring the limits of small language models.
|
| 172 |
|
| 173 |
---
|
|
|
|
| 96 |
3 + 2 = 5
|
| 97 |
</THINK>
|
| 98 |
<ANSWER>
|
| 99 |
+
João tem 5 maçãs.
|
| 100 |
</ANSWER>
|
| 101 |
```
|
| 102 |
|
|
|
|
| 147 |
## 🔮 Future Work
|
| 148 |
|
| 149 |
* Scaling to 10M–50M parameters
|
| 150 |
+
* Improving dataset quality and training techniques
|
| 151 |
* Enhancing reasoning consistency
|
| 152 |
* Multilingual support
|
| 153 |
|
| 154 |
+
|
| 155 |
---
|
| 156 |
|
| 157 |
## 🤝 Contributions
|
| 158 |
|
| 159 |
+
This is an experimental project, contributions and ideas are welcome.
|
| 160 |
|
| 161 |
---
|
| 162 |
|
|
|
|
| 169 |
## 🧠 Author
|
| 170 |
|
| 171 |
AxionLab Co.
|
| 172 |
+
|
| 173 |
Independent research project exploring the limits of small language models.
|
| 174 |
|
| 175 |
---
|