ai.canshu.com
中文
VRAM Calculator Advisor & TCO API Pricing Compare Models Leaderboard GPUs

Search models, GPUs, and tools

↑↓ Navigate · Enter Open · ★ Watchlist prioritized ai.canshu.com
Q4_K_M: Out of VRAM (OOM) 376.18 GB / 99.8 GB Usable

Step 5 Preview (600B MoE) on AMD Ryzen AI Max+ 395 Mini PC (128GB Unified): VRAM Requirement & Speed Estimate

A single AI Max+ 395 128G (99.8 GB usable VRAM) cannot fully fit Step 5 Preview (600B MoE) at Q4_K_M (376.18 GB required). Consider multi-GPU tensor parallelism, a smaller parameter variant, or cloud GPU / API deployment.

Concurrency (Batch):
MoE 604B / A27B·92L (d=4096)·64Q/4KV·Ctx 1000K·Text+Image+Video
Runnable Models →
128 GB Unified·256 GB/s·Safe Usable 99.8 GB
Out of VRAM (OOM)376.18 / 99.8 GBOOM
3. Weight Quantization
4. Per-Request Context16K (16,384 tok)
Verified: 2026-11-20·Roofline bandwidth/compute estimate (Batch=1)
Out of VRAM (OOM)
376.18/ 99.8 GB usable

Physical 128 GB · Load 376.8%

Theoretical Decode Speed
—tokens/s

Exceeds usable VRAM by 276.34 GB

Max Context at Batch=1
0K/ 1000K

Enable Q8/Q4 KV quantization to scale batch concurrency

Weights: 366.18 GBKV Cache (1×16K): 1.38 GBOverhead: 8.62 GB
Reserves ~22% unified memory for OS
⛔ Cannot Load
Reference Launch CLI:
OLLAMA_CONTEXT_LENGTH=16384 ollama run step-5-preview-600b

Insufficient Local VRAM? Ways to Run Step 5 Preview (600B MoE)

Rent an hourly high-VRAM cloud GPU pod, or check the 12-month Buy vs Cloud vs API break-even curve.

📊 All Quantization Levels for Step 5 Preview (600B MoE) on AI Max+ 395 128G

Quantization Total VRAM (16K) Verdict Est. Speed Max Context
FP16 / BF16 1234.84 GB Out of VRAM (OOM) — —
FP8 / Q8_0 657.27 GB Out of VRAM (OOM) — —
GGUF Q6_K 507.87 GB Out of VRAM (OOM) — —
GGUF Q5_K_M 440.87 GB Out of VRAM (OOM) — —
Q4_K_M / AWQ 376.18 GB Out of VRAM (OOM) — —
GGUF IQ4_XS 329.97 GB Out of VRAM (OOM) — —
GGUF Q3_K_M 303.79 GB Out of VRAM (OOM) — —
GGUF Q2_K 245.26 GB Out of VRAM (OOM) — —

💡 Recommended Alternative GPUs & Macs for Step 5 Preview (600B MoE)

Frequently Asked Questions

Can AMD Ryzen AI Max+ 395 Mini PC (128GB Unified) run Step 5 Preview (600B MoE) locally?

A single AI Max+ 395 128G (99.8 GB usable VRAM) cannot fully fit Step 5 Preview (600B MoE) at Q4_K_M (376.18 GB required). Consider multi-GPU tensor parallelism, a smaller parameter variant, or cloud GPU / API deployment.

What is the best quantization for Step 5 Preview (600B MoE) on AI Max+ 395 128G?

Step 5 Preview (600B MoE) (604B parameters) exceeds single-device VRAM even at aggressive 2–3 bit quantization. Cloud GPUs or hosted APIs are recommended.

How much VRAM does Step 5 Preview (600B MoE) need in full FP16 precision?

In unquantized FP16/BF16 precision (16K context), Step 5 Preview (600B MoE) requires 1234.84 GB VRAM (1208 GB for weights alone). Because it uses a Mixture-of-Experts (MoE) architecture, all 604B parameters must sit in memory, while only 27B active parameters are read per token.