Skip to content
HQ
HardwareHQ
Live Pricing

✅ — Radeon Instinct MI325X Can Run Phi-3 Mini 3.8B

The Radeon Instinct MI325X packs 288GB VRAM. Phi-3 Mini 3.8B needs ~2.1GB. You're good to go.

✅ Plenty of headroom

Hardware Specs

GPU
Radeon Instinct MI325X
VRAM
288 GB
Bandwidth
10300 GB/s
MSRP
$0

Model Requirements

Model
Phi-3 Mini 3.8B
Parameters
3.8B
VRAM (Q4)
~2.1 GB
Context
128,000

Estimated Performance

4066 tok/s
Estimated inference speed (Q4 quantization)

Run it locally with Ollama

ollama run phi3:3.8b

Requires Ollama installed on your machine

View on Ollama
0
Phi-3 Mini 3.8B Details Radeon Instinct MI325X Specs

Try Phi-3 Mini 3.8B on a different GPU

Try another model on Radeon Instinct MI325X

Full compatibility matrix

Related Checks

GB200 NVL72 + Phi-3 Mini 3.8B✅AMD Instinct MI350X + Phi-3 Mini 3.8B✅AMD Instinct MI355X + Phi-3 Mini 3.8B✅Radeon Instinct MI350X + Phi-3 Mini 3.8B✅

More tools

VRAM Calculator

Check GPU compatibility for any AI model

Cloud GPU Pricing

Compare pricing across 24+ providers

GPU Comparison

Side-by-side GPU specs and benchmarks

Compatibility Lab

GPU × Model compatibility matrix

HQ
HardwareHQ

AI infrastructure decision engine. Track the GPU market, simulate workloads, and find the most cost-effective setup for any AI project.

Market & Pricing

  • Cloud GPU Pricing
  • Buy vs Rent Calculator
  • GPU Comparison
  • AI Benchmarks
  • Inference Calculator
  • Training Calculator

Planning Studios

  • Research Studio
  • RAG Studio
  • Creative Studio
  • Coding Studio
  • Edge AI Studio
  • Hardware Architect
  • Topology Lab

Reference & Guides

  • VRAM Calculator
  • Model Finder
  • Compatibility Lab
  • LLM Requirements
  • Quantization Guide
  • Best GPU for LLaMA
  • Best GPU for DeepSeek

Company

  • About
  • Contact
  • Advertise
  • Privacy Policy
  • Terms of Service

© 2026 HardwareHQ. All rights reserved.

@HardwareHQio