Can Ryzen AI Max+ 395 128GB run GLM-4.7-Flash?

AMD · 96 GB LPDDR5X · 215 GB/s bandwidth · Z.ai · 31.2B (active 3B) · MoE · model ctx up to 198K

ROCM

Yes — the Ryzen AI Max+ 395 128GB (96 GB LPDDR5X) can run GLM-4.7-Flash. At Q4 with the default 8K context it needs ≈ 18.76 GB VRAM (weights 17.16 GB + KV cache 0.6 GB + 1 GB runtime overhead), which fits within 96 GB. Decode at Q4/8K: ≈ 45.6 tok/s (estimated, batch 1). FP16 (≈ 64.0 GB) also fits, at ≈ 12.5 tok/s (estimated). 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	18.46 GB	Yes	45.6 est.
Q4	8K default	18.76 GB	Yes	45.6 est.
Q4	32K	20.56 GB	Yes	45.6 est.
FP16	4K	63.7 GB	Yes	12.5 est.
FP16	8K default	64.0 GB	Yes	12.5 est.
FP16	32K	65.8 GB	Yes	12.5 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	17.16 GB	0.6 GB	1 GB	18.76 GB	96 GB
FP16	62.4 GB	0.6 GB	1 GB	64.0 GB	96 GB

Other GPUs that run GLM-4.7-Flash

• B200 192GB SXM — ≈1697.0 tok/s Q4 est.
• MI300X 192GB — ≈1124.2 tok/s Q4 est.
• H200 141GB SXM5 — ≈1018.2 tok/s Q4 est.
• H100 80GB SXM5 — ≈710.6 tok/s Q4 est.
• Gaudi 3 128GB — ≈694.9 tok/s Q4 est.
• A100 80GB SXM4 — ≈432.5 tok/s Q4 est.
• RTX 5090 32GB — ≈380.1 tok/s Q4 est.
• RTX 4090 24GB — ≈213.8 tok/s Q4 est.

All GPUs for GLM-4.7-Flash →

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.
• 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.
• Qwen3-Coder-Next — ≈45.6 tok/s Q4 est.

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