Can Mac Studio M3 Ultra 96GB run Qwen3-8B?

Apple · 96 GB LPDDR5 · 819 GB/s bandwidth · Alibaba · 8.2B · model ctx up to 128K

Fit & speed by quant and context

Fits = weights + KV(ctx) + 0.5 GB overhead ≤ 95% of VRAM. Measured anchors are context-agnostic; the fit verdict is recomputed per context. A missing tok/s means the model is far beyond this card (offload-only territory).

VRAM breakdown at 8K context

Other GPUs that run Qwen3-8B

• B200 192GB SXM — ≈916.7 tok/s Q4 est.
• MI300X 192GB — ≈599.4 tok/s Q4 est.
• H200 141GB SXM5 — ≈550.0 tok/s Q4 est.
• H100 80GB SXM5 — ≈383.9 tok/s Q4 est.
• Gaudi 3 128GB — ≈370.5 tok/s Q4 est.
• A100 80GB SXM4 — ≈233.7 tok/s Q4 est.
• RTX 5090 32GB — ≈145.3 tok/s Q4 measured
• RTX 4090D 24GB (China) — ≈105.0 tok/s Q4 est.

All GPUs for Qwen3-8B →

Other models for the Mac Studio M3 Ultra 96GB

• MiniCPM5-1B — ≈419.5 tok/s Q4 est.
• Qwen3.5-2B — ≈226.5 tok/s Q4 est.
• Llama-3.2-3B-Instruct — ≈154.6 tok/s Q4 measured
• Qwen3.6-35B-A3B — ≈151.0 tok/s Q4 est.
• Qwen3.5-35B-A3B — ≈151.0 tok/s Q4 est.
• GLM-4.7-Flash — ≈151.0 tok/s Q4 est.
• Qwen3-30B-A3B-Instruct-2507 — ≈151.0 tok/s Q4 est.
• Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 — ≈151.0 tok/s Q4 est.

All models on the Mac Studio M3 Ultra 96GB →