Is Phi-3.5-MoE-instruct or GPT-Image-2 cheaper?

Phi-3.5-MoE-instruct is cheaper on output tokens ($0.64 vs $30.00 per 1M).

Which has the larger context window, Phi-3.5-MoE-instruct or GPT-Image-2?

Phi-3.5-MoE-instruct has the larger context window (128K tokens).

Phi-3.5-MoE-instruct vs GPT-Image-2

Phi-3.5-MoE-instruct is cheaper on output tokens, while Phi-3.5-MoE-instruct offers a larger context window. Choose Phi-3.5-MoE-instruct or GPT-Image-2 based on the trade-off between cost, context, and the benchmarks that matter for your use case.

Spec	Phi-3.5-MoE-instruct	GPT-Image-2
Provider	Azure	Azure
Input / 1M tokens	$0.16	$5.00
Output / 1M tokens	$0.64	$30.00
Context window	128K	0
Parameters	—	—
Open weights	Yes	No
Released	Aug 2024	Apr 2026

Phi-3.5-MoE-instruct details →GPT-Image-2 details →

FAQ

Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.