AI 模型情報

Phi-4-multimodal

microsoft/phi-4-multimodal

出品方： Microsoft · 系列: phi · 發布 2024-12-11 · 知識截止: 2023-10

$0.080
輸入 / 1M token
$0.320
輸出 / 1M token
128K
上下文長度
4K
最大輸出

Prices in USD per 1M tokens. Unknown means the provider does not publish per-token pricing.

能力清單

✗ 工具呼叫✗ 推理? 結構化輸出✓ 附件✓ 開放權重✓ 溫度可調
支援模態: 輸入 text, image, audio · 輸出 text

Model fit scores

0–100 · higher is better

These scores reward declared capabilities, context size, price and provider availability — they are not benchmark results. Use them as a directional signal alongside your own evaluation.

Coding4
  • Tool calling0/40
  • Structured output0/20
  • Reasoning0/10
  • Context window (100K → 1M)2/20
  • Provider availability2/10
Agents2
  • Tool calling0/35
  • Structured output0/25
  • Reasoning0/15
  • Output token limit0/15
  • Provider availability2/10
JSON / structured output29
  • Structured output / JSON mode0/50
  • Tool calling0/20
  • Temperature control10/10
  • Price-friendly for high-volume19/20
Cost efficiency72
  • Headline price (log-scaled)72/95
  • Has prompt-cache pricing0/5
Long context45
  • Context window (100K → 2M)35/90
  • Has published price for full window10/10
Vision79
  • Accepts image input50/50
  • Context window (10K → 1M)17/30
  • Has published price10/10
  • Provider availability2/10
Production-readiness70
  • Number of independent providers10/40
  • Has published per-token price20/20
  • Context window ≥ 8K15/15
  • No data inconsistencies across providers10/10
  • Official model (not derivative)15/15

Cost Efficiency Index

Open full calculator →

Estimated cost using the recommended provider's headline rate. Each scenario fixes average input/output tokens — the assumptions are shown in the third column.

ScenarioCostAssumption
RAG answer
per 1,000 RAG answers
$0.56
< $0.01 per request
5K input tokens (query + 4 retrieved chunks of ~1K each) and a 500-token answer. Typical SaaS knowledge-base bot.
Support ticket triage
per 10,000 tickets
$1.12
< $0.01 per request
1K input tokens (ticket body + system prompt) and a 100-token JSON classification reply. High-volume customer support.
Data extraction
per 1,000 documents
$0.32
< $0.01 per request
2K input tokens (a single document page) and a 500-token JSON extraction. ETL / invoice / form pipelines.
Code review
per 1,000 PRs
$0.96
< $0.01 per request
8K input tokens (diff + surrounding files) and a 1K-token review comment. PR-bot workloads.
Agent step
per 1,000 steps
$1.15
< $0.01 per request
12K input tokens (long-running tool history) and a 600-token tool-call decision. Cost per agent step.

定價詳情

推薦定價來自 azure · phi-4-multimodal

$0.080
輸入
$0.320
輸出
$4.00
輸入音訊

於 2 家供應商可用

服務商服務商模型 ID輸入 / 1M輸出 / 1M上下文發布日期
Azure
azure
phi-4-multimodal$0.080$0.320128K2024-12-11
Azure Cognitive Services
azure-cognitive-services
phi-4-multimodal$0.080$0.320128K2024-12-11

Frequently asked questions

How much does Phi-4-multimodal cost?

Phi-4-multimodal costs $0.080 per 1M input tokens and $0.320 per 1M output tokens, sourced from azure. Cache reads, audio tokens and >200K-context tiers (where applicable) are listed in the Pricing detail block above.

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

Phi-4-multimodal has a context window of 128K tokens, with a max output of 4K tokens per reply. This is the total combined size of prompt + completion.

Does Phi-4-multimodal support tool calling?

No. Phi-4-multimodal does not support tool calling (function calling). If your workflow requires it, look at the /capabilities/tool-calling list for alternatives.

Does Phi-4-multimodal support structured output / JSON mode?

Support for structured output / JSON-schema-constrained decoding is not reported for Phi-4-multimodal in our data source. Verify with Microsoft's official documentation before relying on it in production.

Can Phi-4-multimodal accept image input?

Yes. Phi-4-multimodal accepts both text and image input. Vision pricing per image is usually billed on top of the regular token rate — check Microsoft's docs for the exact rule.

Is Phi-4-multimodal open-weight?

Yes. Phi-4-multimodal's weights are publicly available, so you can self-host or fine-tune. Note that open weights ≠ open source — the training data and code are typically not released.

What are the best alternatives to Phi-4-multimodal?

If Phi-4-multimodal doesn't fit, consider Phi-4-mini-instruct, Phi-4, WizardLM-2 8x22B. Each one targets the same use case — see the Related links below for direct head-to-head pages.

Where does this data come from?

All numbers come from the public models.dev API and are normalised into a single canonical model record. We re-pull daily and write any changes (price, context, capability) to the /changelog page.

Explore more

More Microsoft models

Capability lists this model is in

See also

最近更新：

Data is sourced from models.dev and normalized for comparison. Prices and capabilities may change. Always verify critical production decisions with the provider's official documentation.