Can Ryzen AI Max+ 395 128GB run Llama-4-Scout-17B-16E-Instruct?

AMD · 96 GB LPDDR5X · 215 GB/s bandwidth · Meta · 109.0B (active 17.0B) · MoE · model ctx up to 10240K

ROCM

Yes — the Ryzen AI Max+ 395 128GB (96 GB LPDDR5X) can run Llama-4-Scout-17B-16E-Instruct. At Q4 with the default 8K context it needs ≈ 64.35 GB VRAM (weights 59.95 GB + KV cache 3.4 GB + 1 GB runtime overhead), which fits within 96 GB. Decode at Q4/8K: ≈ 8.0 tok/s (estimated, batch 1). FP16 needs ≈ 222.4 GB — does not fit on this card. Estimates come from memory-bandwidth math; rows tagged measured override estimates. Relative ranking is more reliable than absolute tok/s. Methodology.

Fit & speed by quant and context

Quant	Context	VRAM needed	Fits?	tok/s (decode)
Q4	4K	62.65 GB	Yes	8.0 est.
Q4	8K default	64.35 GB	Yes	8.0 est.
Q4	32K	74.55 GB	Yes	8.0 est.
FP16	4K	220.7 GB	No	1.2 est.
FP16	8K default	222.4 GB	No	1.2 est.
FP16	32K	232.6 GB	No	1.1 est.

Fits = weights + KV(ctx) + 1 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

Quant	Weights	KV cache	Overhead	Total needed	Ryzen AI Max+ 395 128GB VRAM
Q4	59.95 GB	3.4 GB	1 GB	64.35 GB	96 GB
FP16	218.0 GB	3.4 GB	1 GB	222.4 GB	96 GB

Other GPUs that run Llama-4-Scout-17B-16E-Instruct

• B200 192GB SXM — ≈299.5 tok/s Q4 est.
• MI300X 192GB — ≈198.4 tok/s Q4 est.
• H200 141GB SXM5 — ≈179.7 tok/s Q4 est.
• H100 80GB SXM5 — ≈125.4 tok/s Q4 est.
• Gaudi 3 128GB — ≈122.6 tok/s Q4 est.
• A100 80GB SXM4 — ≈76.3 tok/s Q4 est.
• Mac Studio M3 Ultra 96GB — ≈30.7 tok/s Q4 est.
• Mac Studio M2 Ultra 192GB — ≈29.9 tok/s Q4 est.

All GPUs for Llama-4-Scout-17B-16E-Instruct →

Other models for the Ryzen AI Max+ 395 128GB

• MiniCPM5-1B — ≈126.7 tok/s Q4 est.
• Qwen3.5-2B — ≈68.4 tok/s Q4 est.
• DeepSeek-R1-0528-Qwen3-8B-layer-mix-bpw-3.8-mlx — ≈59.5 tok/s Q4 est.
• Qwen3.6-35B-A3B — ≈45.6 tok/s Q4 est.
• Qwen3.5-35B-A3B — ≈45.6 tok/s Q4 est.
• GLM-4.7-Flash — ≈45.6 tok/s Q4 est.
• Qwen3-30B-A3B-Instruct-2507 — ≈45.6 tok/s Q4 est.
• Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 — ≈45.6 tok/s Q4 est.

All models on the Ryzen AI Max+ 395 128GB →