LLM Router
HomeRoutersModelsProvidersBenchmarksPricingCompareBlogAbout
HomeModelsBenchmarksPricingCompareBlog
LLM Router

Independent comparison platform for LLM routing infrastructure.

Platform

  • Home
  • Routers
  • Models
  • Pricing
  • Blog
  • About

Routers

  • Requesty
  • OpenRouter
  • Martian
  • Unify
  • LiteLLM

© 2026 LLM Router

Data from public sources. May not reflect real-time pricing.

Providers›Microsoft

Microsoft

↗ Website

Microsoft Research develops the Phi family of small but capable language models. The Phi models demonstrate that smaller models can achieve impressive performance through high-quality training data and innovative techniques.

Pricing available from Requesty, Martian, OpenRouter, DeepInfra.

Total Models
2
Arena Ranked
1
of 2
Open Source
2
of 2
Cheapest Input
$0.06
per 1M tokens

$ Pricing Summary(per 1M tokens)

MetricInputOutput
Cheapest$0.06$0.14
Average$0.06$0.14
Most Expensive$0.07$0.14

⚙ Capabilities

👁
Vision
0
of 2 models
🧠
Reasoning
0
of 2 models
🔧
Tool Calling
0
of 2 models
⚡
Prompt Caching
0
of 2 models
🖥
Computer Use
0
of 2 models
🎨
Image Generation
0
of 2 models

🤖 All Microsoft Models(2)

MicrosoftPhi 4OSS
#209

Microsoft: Phi 4

[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed. At 14 billion parameters, it was trained on a mix of high-quality synthetic datasets, data from curated websites, and academic materials. It has undergone careful improvement to follow instructions accurately and maintain strong safety standards. It works best with English language inputs. For more information, please see [Phi-4 Technical Report](https://arxiv.org/pdf/2412.08905)

Context
16K
Max Output
16K
Input/1M
$0.06
Pricing (per 1M tokens)
OpenRouter$0.06 / $0.14
Martian$0.06 / $0.14
DeepInfra$0.07 / $0.14
2025-01-10View details →
MicrosoftPhi 4OSS

Phi 4

Phi-4-reasoning-plus is an enhanced 14B parameter model from Microsoft, fine-tuned from Phi-4 with additional reinforcement learning to boost accuracy on math, science, and code reasoning tasks. It uses the same dense decoder-only transformer architecture as Phi-4, but generates longer, more comprehensive outputs structured into a step-by-step reasoning trace and final answer. While it offers improved benchmark scores over Phi-4-reasoning across tasks like AIME, OmniMath, and HumanEvalPlus, its responses are typically ~50% longer, resulting in higher latency. Designed for English-only applications, it is well-suited for structured reasoning workflows where output quality takes priority over response speed.

Context
16K
Max Output
—
Input/1M
$0.06
Pricing (per 1M tokens)
Requesty★$0.07 / $0.14
Martian$0.06 / $0.14
View details →
← Back to all providers