Compatible devices for Phi-3.5 Mini (3.8B)
Every hardware profile in our catalog graded for Phi-3.5 Mini (3.8B), best fit first. For sellable vendor configurations, see the device catalog.
- Supermicro 8x H100 SuperServerSupermicro · AI Servers
Fits at FP16 (~8GB) with ~555.2GB headroom — about 70 concurrent instances.
FP16 · ~8GBRuns well - Dell PowerEdge XE9680Dell · AI Servers
Fits at FP16 (~8GB) with ~555.2GB headroom — about 70 concurrent instances.
FP16 · ~8GBRuns well - AMD Instinct MI300XAMD · Datacenter GPUs
Fits at FP16 (~8GB) with ~161GB headroom — about 21 concurrent instances.
FP16 · ~8GBRuns well - NVIDIA H200 (141GB)NVIDIA · Datacenter GPUs
Fits at FP16 (~8GB) with ~116.1GB headroom — about 15 concurrent instances.
FP16 · ~8GBRuns well - Cloud H200 141GB (profile)Cloud · Cloud GPU Profiles
Fits at FP16 (~8GB) with ~116.1GB headroom — about 15 concurrent instances.
FP16 · ~8GBRuns well - NVIDIA H100 (80GB)NVIDIA · Datacenter GPUs
Fits at FP16 (~8GB) with ~62.4GB headroom — about 8 concurrent instances.
FP16 · ~8GBRuns well - Cloud H100 80GB (profile)Cloud · Cloud GPU Profiles
Fits at FP16 (~8GB) with ~62.4GB headroom — about 8 concurrent instances.
FP16 · ~8GBRuns well - HP Z8 Fury G5 WorkstationHP · AI Workstations
Fits at FP16 (~8GB) with ~76.5GB headroom — about 10 concurrent instances.
FP16 · ~8GBRuns well - Lenovo ThinkStation PX WorkstationLenovo · AI Workstations
Fits at FP16 (~8GB) with ~76.5GB headroom — about 10 concurrent instances.
FP16 · ~8GBRuns well - Supermicro AI WorkstationSupermicro · AI Workstations
Fits at FP16 (~8GB) with ~76.5GB headroom — about 10 concurrent instances.
FP16 · ~8GBRuns well - Apple Mac Studio (M2 Ultra)Apple · Apple Silicon
Fits at FP16 (~8GB) with ~126.4GB headroom — about 16 concurrent instances.
FP16 · ~8GBRuns well - Quad RTX 4090 AI Workstation (reference profile)Reference · AI Workstations
Fits at FP16 (~8GB) with ~76.5GB headroom — about 10 concurrent instances.
FP16 · ~8GBRuns well - Dell Precision 7960 AI WorkstationDell · AI Workstations
Fits at FP16 (~8GB) with ~34.2GB headroom — about 5 concurrent instances.
FP16 · ~8GBRuns well - NVIDIA A100 80GBNVIDIA · Datacenter GPUs
Fits at FP16 (~8GB) with ~62.4GB headroom — about 8 concurrent instances.
FP16 · ~8GBRuns well - Cloud A100 80GB (profile)Cloud · Cloud GPU Profiles
Fits at FP16 (~8GB) with ~62.4GB headroom — about 8 concurrent instances.
FP16 · ~8GBRuns well - Apple Mac Studio (M4 Max)Apple · Apple Silicon
Fits at FP16 (~8GB) with ~81.6GB headroom — about 11 concurrent instances.
FP16 · ~8GBRuns well - NVIDIA DGX Spark (GB10)NVIDIA · AI Appliances
Fits at FP16 (~8GB) with ~81.6GB headroom — about 11 concurrent instances.
FP16 · ~8GBRuns well - ASUS Ascent GX10 (GB10)ASUS · AI Appliances
Fits at FP16 (~8GB) with ~81.6GB headroom — about 11 concurrent instances.
FP16 · ~8GBRuns well - Dell Pro Max with GB10Dell · AI Appliances
Fits at FP16 (~8GB) with ~81.6GB headroom — about 11 concurrent instances.
FP16 · ~8GBRuns well - AMD Ryzen AI Max Mini PC (Strix Halo class)AMD · Mini PCs
Fits at FP16 (~8GB) with ~81.6GB headroom — about 11 concurrent instances.
FP16 · ~8GBRuns well - Coding Agent Workstation (reference profile)Reference · AI Workstations
Fits at FP16 (~8GB) with ~34.2GB headroom — about 5 concurrent instances.
FP16 · ~8GBRuns well - NVIDIA L40SNVIDIA · Datacenter GPUs
Fits at FP16 (~8GB) with ~34.2GB headroom — about 5 concurrent instances.
FP16 · ~8GBRuns well - Cloud L40S 48GB (profile)Cloud · Cloud GPU Profiles
Fits at FP16 (~8GB) with ~34.2GB headroom — about 5 concurrent instances.
FP16 · ~8GBRuns well - Apple Mac mini (M4 Pro)Apple · Apple Silicon
Fits at FP16 (~8GB) with ~36.8GB headroom — about 5 concurrent instances.
FP16 · ~8GBRuns well - Law Firm Private AI Box (reference profile)Reference · AI Appliances
Fits at FP16 (~8GB) with ~34.2GB headroom — about 5 concurrent instances.
FP16 · ~8GBRuns well - NVIDIA RTX 6000 Ada GenerationNVIDIA · Professional GPUs
Fits at FP16 (~8GB) with ~34.2GB headroom — about 5 concurrent instances.
FP16 · ~8GBRuns well - AMD Radeon PRO W7900AMD · Professional GPUs
Fits at FP16 (~8GB) with ~34.2GB headroom — about 5 concurrent instances.
FP16 · ~8GBRuns well - NVIDIA RTX A6000NVIDIA · Professional GPUs
Fits at FP16 (~8GB) with ~34.2GB headroom — about 5 concurrent instances.
FP16 · ~8GBRuns well - Accounting / Odoo AI Box (reference profile)Reference · AI Appliances
Fits at FP16 (~8GB) with ~13.1GB headroom — about 2 concurrent instances.
FP16 · ~8GBRuns well - Small Business Mini PC (reference profile)Reference · Mini PCs
Fits at FP16 (~8GB) with ~14.4GB headroom — about 2 concurrent instances.
FP16 · ~8GBRuns well - NVIDIA GeForce RTX 4090NVIDIA · Consumer GPUs
Fits at FP16 (~8GB) with ~13.1GB headroom — about 2 concurrent instances.
FP16 · ~8GBRuns well - Apple Mac mini (M4)Apple · Apple Silicon
Fits at FP16 (~8GB) with ~14.4GB headroom — about 2 concurrent instances.
FP16 · ~8GBRuns well - AMD Radeon RX 7900 XTXAMD · Consumer GPUs
Fits at FP16 (~8GB) with ~13.1GB headroom — about 2 concurrent instances.
FP16 · ~8GBRuns well - NVIDIA GeForce RTX 3090NVIDIA · Consumer GPUs
Fits at FP16 (~8GB) with ~13.1GB headroom — about 2 concurrent instances.
FP16 · ~8GBRuns well - Dual RTX 3060 Local Server (reference profile)Reference · AI Servers
Fits at FP16 (~8GB) with ~13.1GB headroom — about 2 concurrent instances.
FP16 · ~8GBRuns well - Local Office AI Appliance (reference profile)Reference · AI Appliances
Fits at FP16 (~8GB) with ~6.1GB headroom — about 1 concurrent instance.
FP16 · ~8GBRuns well - Hotel AI Automation Box (reference profile)Reference · AI Appliances
Fits at FP16 (~8GB) with ~6.1GB headroom — about 1 concurrent instance.
FP16 · ~8GBRuns well - Intel Arc A770 16GBIntel · Consumer GPUs
Fits at FP16 (~8GB) with ~6.1GB headroom — about 1 concurrent instance.
FP16 · ~8GBRuns well - Intel Arc B580 12GBIntel · Consumer GPUs
Fits at FP16 (~8GB) with ~2.6GB headroom — about 1 concurrent instance.
FP16 · ~8GBRuns well - NVIDIA GeForce RTX 3060 12GBNVIDIA · Consumer GPUs
Fits at FP16 (~8GB) with ~2.6GB headroom — about 1 concurrent instance.
FP16 · ~8GBRuns well
Run Phi-3.5 Mini (3.8B) privately
Run your own AI agents on hardware you control — private by design, no per-seat data leaving your premises. BrainOutput helps you pick the right machine and turn it into a working AI Business OS.