HuggingFaceTB/SmolLM2-135M
🧠 AI ModelHuggingFaceTB
A high-performance, ultra-compact 135M parameter language model optimized for edge devices and resource-constrained environments.
SmolLM2-135M represents a breakthrough in small language model (SLM) engineering. Built by HuggingFaceTB, this model utilizes the Llama architecture to deliver robust text-generation capabilities despite its extremely small parameter count. It is specifically optimized for low-latency inference, making it ideal for real-time applications on hardware with limited RAM or processing power. The model is released under the Apache 2.0 license, ensuring broad accessibility for both research and commercial deployment. It supports standard Hugging Face transformers workflows, allowing for seamless integration into existing pipelines. Its architecture is fine-tuned to maximize performance on English language tasks, demonstrating that high-quality synthetic data and efficient training recipes can produce highly capable models that punch well above their weight class in terms of parameter efficiency.
💡Highlights
- ├─135M parameter architecture
- ├─Optimized for edge deployment
- └─Apache 2.0 open license
🎯For
- ├─Edge AI Developers
- ├─Mobile App Engineers
- └─Embedded Systems Researchers