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...

Context window: 16,384 tokens
Max output: 14,745 tokens
Reasoning effort: Not adjustable
Tool calls: No
Temperature: Adjustable

Providers and prices

Providers serving Phi 4 with prices per million tokens
Provider	Quantization	Input / M tokens	Output / M tokens
DeepInfradeepinfra/bf16/flex	bf16	$0.056	$0.112
DeepInfradeepinfra/bf16	bf16	$0.07	$0.14

Prices are what the Provider charges, synced 2026-10-21. Amber adds its platform fee to each Trial’s total; see pricing.

Model Configurations

Amber has no curated Model Configuration for this Model yet. Create a custom one in your Workspace to evaluate it.

Works with these Agents

Terminus-2
pi
mini-swe-agent
Direct

All Models

Model