 Phi-3.5 Mini 128K Instruct | Phi 4 Multimodal Instruct | |
|---|---|---|
Model Provider The organization behind this AI's development | Microsoft | Microsoft |
Input Context Window Maximum input tokens this model can process at once | 128K tokens | 131.1K tokens |
Output Token Limit Maximum output tokens this model can generate at once | Not specified tokens | Not specified tokens |
Release Date When this model first became publicly available | August 21, 2024 1 year ago August 21st, 2024 | March 8, 2025 8 months ago March 8th, 2025 |
Phi-3.5 Mini 128K Instruct | Phi 4 Multimodal Instruct | |
|---|---|---|
Input Types Supported input formats | 📝Text | 📝Text🖼️Image |
Output Types Supported output formats | 📝Text | 📝Text |
Tokenizer Text encoding system | Other | Other |
Key Features Advanced capabilities | ✓Function CallingStructured OutputReasoning ModeContent Moderation | Function Calling✓Structured OutputReasoning ModeContent Moderation |
Open Source Model availability | Available on HuggingFace → | Available on HuggingFace → |
| Phi-3.5 Mini 128K Instruct | Phi 4 Multimodal Instruct | |
|---|---|---|
Input Token Cost Cost per million input tokens | $0.10 per million tokens | $0.05 per million tokens |
Output Token Cost Cost per million outut tokens | $0.10 per million tokens | $0.10 per million tokens |
| Phi-3.5 Mini 128K Instruct | Phi 4 Multimodal Instruct | |
|---|---|---|
MMLU Measures knowledge across 57 subjects like law, math, history, and science | Benchmark not available. | Benchmark not available. |
MMMU Measures understanding of combined text and images across various domains | Benchmark not available. | Benchmark not available. |
HellaSwag Measures common sense reasoning by having models complete sentences about everyday situations | Benchmark not available. | Benchmark not available. |
Phi-3.5 Mini 128K Instruct by Microsoft can use external tools and APIs. It can handle standard conversations with its 128K token context window. Very affordable at $0.10/M input and $0.10/M output tokens. Released August 21st, 2024.Phi 4 Multimodal Instruct by Microsoft understands both text and images, generates structured data. It can handle standard conversations with its 131.1K token context window. Very affordable at $0.05/M input and $0.10/M output tokens. Released March 8th, 2025.Compare your models with top performers across different categories
Best models for coding and development
Models optimized for creative writing
Content creation and marketing tasks
Technical analysis and explanations
Scientific research and analysis
Multilingual translation tasks
Best models for coding and development
Models optimized for creative writing
Content creation and marketing tasks
Technical analysis and explanations
Scientific research and analysis
Multilingual translation tasks
