Apple M3 Max (30-core GPU) — Local LLM Performance & Compatibility

Written by Jakub Rusinowski · Last updated September 19, 2026

The 14-core CPU / 30-core GPU M3 Max. Fewer memory controllers than the 40-core part: 300 GB/s rather than 400, and only 36 GB or 96 GB configurations. Modelling a 36 GB machine on the 40-core record overstates its throughput by a third.

Technical Specifications

VRAM96 GB unified memory
Memory Bandwidth300 GB/s
TDP35 W
ArchitectureARM, 3nm TSMC
Release Year2023
MSRP at Launch$0
Inference Speed (Llama 3.1 8B Q4_K_M)18–38 tok/s (estimated)
Inference Speed (Llama 3.3 70B Q4_K_M)2.4–5.0 tok/s (estimated)
Buy This HardwareRyzen AI Max+ 395 Laptop (Strix Halo, up to 128GB) — 128 GB VRAM · 120 W board powerDeploy in the Cloud NowRunPod

or compare on Vast.ai

As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.

LLMs Compatible with 96 GB Unified Memory

All models below run comfortably in 96 GB unified memory with Q4_K_M quantization.

Qwen 3.5Qwen 3.5 122B-A10B · 74 GB VRAM · Q4_K_M · ollama run qwen3.5:122b
Mistral Small 4Mistral Small 4 119B-A6.5B · 73 GB VRAM · Q4_K_M · ollama run mistral-small
Llama 4Llama 4 Scout 17B · 67 GB VRAM · Q4_K_M · ollama run llama4:scout
Command R FamilyCommand R+ (104B) · 64 GB VRAM · Q4_K_M · ollama run command-r-plus
Llama 3.2 FamilyLlama 3.2 90B Vision Instruct · 54 GB VRAM · Q4_K_M · llama-3-2
Llama 3.2 VisionLlama 3.2 Vision 90B · 54 GB VRAM · Q4_K_M · ollama run llama3.2-vision:90b
Qwen3-CoderQwen3-Coder 80B-A3B (MoE) · 49 GB VRAM · Q4_K_M · ollama run qwen3-coder:80b-a3b-q4
Qwen 2.5 VLQwen 2.5 VL 72B Instruct · 45 GB VRAM · Q4_K_M · ollama run qwen2.5vl:72b

62 more families also fit 96 GB — browse the full model library.

Best Use Cases

FAQ

Can the Apple M3 Max (30-core GPU) run local LLMs?

Yes — the Apple M3 Max (30-core GPU) has 96 GB unified memory and runs The 14-core CPU / 30-core GPU M3 Max. Fewer memory controllers than the 40-core part: 300 GB/s rather than 400, and only

How fast is the Apple M3 Max (30-core GPU) for AI inference?

The Apple M3 Max (30-core GPU) is estimated to run Llama 3.1 8B at 18–38 tok/s with Q4_K_M quantization. For Llama 3.3 70B the estimate is 2.4–5.0 tok/s. These are modelled estimates, not measurements — see /en/methodology.

What LLMs can I run on 96 GB VRAM?

With 96 GB you can run: Qwen 3.5, Mistral Small 4, Llama 4, Command R Family, Llama 3.2 Family. Use Ollama for the easiest setup: ollama run llama3.1:8b.

Compare Similar GPUs

VRAM Tier

Buying Guide

← All GPU Reviews | All Hardware | Check Your Hardware | Full Benchmarks | Can I Run It?