Can you run Qwen 3.8 Flash (Flash-Next) locally?

qwen family · Sep 2026 · 125B parameters (6B activated per token) · Hugging Face model card

How much VRAM does Qwen 3.8 Flash (Flash-Next) need at each quantization?

Qwen 3.8 Flash (Flash-Next) needs 82.5 GB of VRAM at Q4_K_M. The table below lists weights-only size and total VRAM including overhead for each common quantization level with 4k and 32k token contexts.

Method: weights = parameters × bits-per-weight (Q4_K_M ≈ 4.8, Q5_K_M ≈ 5.7, Q6_K ≈ 6.6, Q8_0 ≈ 8.5, FP16 = 16), plus 10% loading overhead, plus KV cache from the verified architecture config (not applicable to this architecture). Source: Qwen3.8-Flash-Next: 125B main (+51B n-gram + 4B MTP, ~177B checkpoint), 6B active/token, hybrid GDN/QSA, 262K context. Sources: qwen.ai blog, Unsloth docs | run2 2026-09-12: repo (Qwen3.8-Flash-Next) safetensors total 180B likely spans multiple formats; param count per Qwen announcement; license 'other' per repo card

Which GPUs can run Qwen 3.8 Flash (Flash-Next) locally?

At Q4_K_M with a 4k context, Qwen 3.8 Flash (Flash-Next) needs 82.5 GB of VRAM. The lists below are computed live from our GPU database and grouped by how much headroom the card has. Disclosure: CompareAIHardware.com participates in the Amazon Associates program and earns from qualifying purchases through links on this page. Affiliate relationships do not influence our recommendations.

Runs Qwen 3.8 Flash (Flash-Next) comfortably (25%+ VRAM headroom)

Minimum GPUs that fit Qwen 3.8 Flash (Flash-Next) at Q4_K_M

These GPUs hold the model but leave little headroom — keep contexts short.

Needs 2+ GPUs to run Qwen 3.8 Flash (Flash-Next)

One of these cards is too small on its own, but a pair (tensor or pipeline parallel, ~90% efficiency) covers the 82.5 GB requirement. See our multi-GPU guide for setup.

Can a Mac run Qwen 3.8 Flash (Flash-Next)?

Yes — these Apple Silicon machines fit Qwen 3.8 Flash (Flash-Next) at Q4_K_M, since macOS lets the GPU use about 75% of unified memory. Generation speed is bound by memory bandwidth, so the GB/s column matters as much as capacity.

Can Qwen 3.8 Flash (Flash-Next) run on mini PCs, Jetson, or NPU devices?

These edge and NPU devices from our database have enough memory for Qwen 3.8 Flash (Flash-Next) at Q4_K_M. Their memory bandwidth is far below discrete GPUs, so expect a fraction of desktop generation speed.

Which pre-built systems can run Qwen 3.8 Flash (Flash-Next)?

These mini PCs and workstations from our database fit Qwen 3.8 Flash (Flash-Next) at Q4_K_M. Usable-memory figures are conservative: Windows shares about half of system RAM with the GPU by default (Linux can expose more), macOS lets Apple Silicon GPUs use about 75% of unified memory, and Linux unified-memory systems such as GB10 expose roughly 90%. Generation speed is bound by memory bandwidth, so compare the GB/s column before buying.

Frequently asked questions