Activefastllm Open Source

Phi-3 Small

by Microsoft· Released May 2024· Cutoff May 2024

Phi-3 Small is a 7B parameter language model from Microsoft's Phi-3 family, designed for efficient performance on edge devices and cloud inference. It achieves strong results on benchmarks while being small enough to run on smartphones, making it a cost-effective alternative to larger models.

Official Site API Docs 🤗 Hugging Face 📄 Research Paper

Input cost

Free (open source)

Output cost

Free (open source)

Context window

128K tokens

Max output

—

Modalities

text

Parameters

License

MIT

Capabilities

Code GenerationReasoningText GenerationFunction CallingStreamingJSON Mode

Best For

On-device and edge deployments requiring strong reasoning and code capabilities with low latency.

Strengths

Small size (7B) enables deployment on mobile and edge devices
Strong reasoning and coding performance relative to size
128K token context window
Open source with MIT license

Limitations

Not suitable for complex multi-step reasoning tasks compared to larger models
Limited multilingual support
No vision or multimodal capabilities
Smaller knowledge base due to fewer parameters

Use Cases

On-device AI assistants

Code completion and generation in IDEs

Text summarization on mobile

Customer support chatbots

Educational tutoring apps

Real-time language translation

Edge IoT analytics

Improvements Over Previous Model

Introduced as a new smaller model in the Phi-3 family, offering a balance between size and performance
128K context window, larger than Phi-2's 2K
MIT license, more permissive than Phi-2's research-only license
Improved reasoning and coding benchmarks over Phi-2
Optimized for on-device deployment with ONNX runtime support

Back to all models

Activefastllm Open Source

Phi-3 Small

by Microsoft· Released May 2024· Cutoff May 2024

Official Site API Docs 🤗 Hugging Face 📄 Research Paper

Input cost

Free (open source)

Output cost

Free (open source)

Context window

128K tokens

Max output

—

Modalities

text

Parameters

License

MIT

Capabilities

Code GenerationReasoningText GenerationFunction CallingStreamingJSON Mode

Best For

On-device and edge deployments requiring strong reasoning and code capabilities with low latency.

Strengths

Small size (7B) enables deployment on mobile and edge devices
Strong reasoning and coding performance relative to size
128K token context window
Open source with MIT license

Limitations

Not suitable for complex multi-step reasoning tasks compared to larger models
Limited multilingual support
No vision or multimodal capabilities
Smaller knowledge base due to fewer parameters

Use Cases

On-device AI assistants

Code completion and generation in IDEs

Text summarization on mobile

Customer support chatbots

Educational tutoring apps

Real-time language translation

Edge IoT analytics

Improvements Over Previous Model

Introduced as a new smaller model in the Phi-3 family, offering a balance between size and performance
128K context window, larger than Phi-2's 2K
MIT license, more permissive than Phi-2's research-only license
Improved reasoning and coding benchmarks over Phi-2
Optimized for on-device deployment with ONNX runtime support

Back to all models