Comparisons of Tiny Open Source AI Models (<4B)
Open source AI models with 4 billion parameters or fewer. These are usually the smallest models in terms of resource demand. Models are considered Open Source (also commonly referred to as open weights) where their weights are accessible to download. This allows self-hosting on your own infrastructure and enables customizing the model such as through fine-tuning. Click on any model to see detailed metrics. For more details including relating to our methodology, see our FAQs.
Qwen3 1.7B (Reasoning) and 
Phi-4 Mini are the highest intelligence Tiny open source models, defined as those with <4B parameters, followed by Phi-3 Mini & Qwen3 1.7B.
Highlights
Intelligence
Artificial Analysis Intelligence Index; Higher is better
Total Parameters
Trainable parameters in billions
Further details
| Weights | Provider Benchmarks | |||||||
|---|---|---|---|---|---|---|---|---|
Qwen3 1.7B (Reasoning) Alibaba | 25 | 2.03B | 32.0k | $0.4 | - | 🤗 | | View |
Phi-4 Mini Instruct Microsoft Azure | 19 | 3.84B | 128k | - | 47 | 🤗 | | View |
Phi-3 Mini Instruct 3.8B Microsoft Azure | 16 | 3.8B | 4.10k | $0.2 | 69 | 🤗 | | View |
Qwen3 1.7B Alibaba | 15 | 2.03B | 32.0k | $0.2 | - | 🤗 | | View |
Qwen3 0.6B (Reasoning) Alibaba | 15 | 0.752B | 32.0k | $0.4 | - | 🤗 | | View |
 Llama 3.2 Instruct 3B Meta | 14 | 3B | 128k | $0.0 | 115 | 🤗 |    +4 more | View |
Qwen3 0.6B Alibaba | 12 | 0.752B | 32.0k | $0.2 | - | 🤗 | | View |
 DeepSeek R1 Distill Qwen 1.5B DeepSeek | 11 | 1.5B | 128k | - | - | 🤗 | - | View |
 LFM2 1.2B Liquid AI | 11 | 1.17B | 32.8k | - | - | 🤗 | ? | View |
Llama 3.2 Instruct 1B Meta | 10 | 1B | 128k | $0.1 | 120 | 🤗 |   | View |
 Gemma 3 1B Instruct Google | 9 | 1B | 32.0k | - | - | 🤗 | ? | View |