Phi-4-multimodal-instruct

by Microsoft

Phi-4-multimodal-instruct is a model from Microsoft, released on February 1, 2025. It has a 128K-token context window and can generate up to 128K output tokens. API pricing is $0.05 per 1M input tokens and $0.1 per 1M output tokens. It ranks #183 of 220 models by Intelligence Score with a score of 2.8.

Phi-4-multimodal-instruct is a lightweight (5.57B parameters) open multimodal foundation model that leverages research and datasets from Phi-3.5 and 4.0. It processes text, image, and audio inputs to generate text outputs, supporting a 128K token context length. Enhanced via SFT, DPO, and RLHF for instruction following and safety.

Key specs

ProviderMicrosoft
Release date2025-02-01
Licenseopen source
Context window128K tokens
Max output128K tokens
Input price$0.05 / 1M tokens
Output price$0.1 / 1M tokens
Intelligence Score2.8

Quick answers

What is the context window of Phi-4-multimodal-instruct?

Phi-4-multimodal-instruct has a context window of 128K tokens, with a maximum output of 128K tokens per response.

How much does Phi-4-multimodal-instruct cost?

Phi-4-multimodal-instruct costs $0.05 per 1M input tokens and $0.1 per 1M output tokens through the API.

Is Phi-4-multimodal-instruct open source?

Yes. Phi-4-multimodal-instruct is an open model released under the open source license.

When was Phi-4-multimodal-instruct released?

Microsoft released Phi-4-multimodal-instruct on February 1, 2025.

How does Phi-4-multimodal-instruct compare to Claude Opus 5.5 (Reasoning)?

Phi-4-multimodal-instruct scores 2.8 on the Intelligence Score versus 60.5 for Claude Opus 5.5 (Reasoning), the current #1. Input pricing is $0.05 vs $4 per 1M tokens.

Compare on the AI leaderboard · All AI models

More models from Microsoft