RunsOnMy ModelsGPUs & MacsBy VRAMBest GPUCan I run…?API pricesVRAM calculatorNew

Can Qwen3.8 Flash Next run on the RTX 4070 Ti?

Not fully in VRAM. Even at 2-bit with 8K context, Qwen3.8 Flash Next needs about 63.9 GB, more than the 11.4 GB usable on the RTX 4070 Ti. It can still run with offload (slower): Q4_K_M with part of the model in system RAM, about 0.7–1.4 tokens/sec (estimate). Details and command.

Runs with offload (slower)

Runs with offload 4-bit (Q4_K_M) at 8K context: 4 of 48 layers on the RTX 4070 Ti (about 9.4 GB) and the other 44 in system RAM (about 94.1 GB). Needs a PC with at least 128 GB of system RAM. Estimated speed: 0.7–1.4 tokens/sec (very slow: fine for batch jobs, painful for chat), versus a bandwidth ceiling of ~13 tok/s if it fit fully in VRAM.

llama-server -m qwen3-8-flash-next-Q4_K_M.gguf -c 8192 -ngl 4

Estimate, not a benchmark. The GGUF repo we found (unsloth/Qwen3.8-Flash-Next-GGUF) has no Q4_K_M file; point -m at your own Q4_K_M GGUF. The speed range spans dual-channel DDR4-3200 to DDR5-6000 system RAM; your CPU, RAM speed and llama.cpp version move it. If llama.cpp runs out of GPU memory, lower -ngl by 2-3; if VRAM is left over, raise it for more speed. Recent llama.cpp builds can also pick the split automatically. How we estimate offload.

Memory by quant and context

Quant4K ctx8K ctx32K ctx128K ctxMax tok/s (est., 8K)
2-bit (Q2_K)63.8 GB No63.9 GB No64.5 GB No66.7 GB No—
3-bit (Q3_K_M)82.9 GB No83.0 GB No83.6 GB No85.8 GB No—
4-bit (Q4_K_M)103 GB No104 GB No104 GB No106 GB No—
5-bit (Q5_K_M)120 GB No121 GB No121 GB No123 GB No—
6-bit (Q6_K)138 GB No139 GB No139 GB No141 GB No—
8-bit (Q8_0) •176 GB No176 GB No177 GB No179 GB No—
16-bit (F16/BF16) •331 GB No331 GB No331 GB No334 GB No—

• real GGUF file size from unsloth/Qwen3.8-Flash-Next-GGUF. Speed is a bandwidth ceiling (504 GB/s), not a benchmark. Methodology.

Hardware that does fit it

A100 80GBH100 PCIeH100 SXMRTX PRO 6000 BlackwellMac M2 Max 96GBDGX Spark

What fits instead

"Runs" and "fits" mean the whole model in GPU memory at 8K context (4-bit or better for the model pick). Prices are US launch MSRPs, not today's street prices. "Popular" means the 60 most-downloaded models we track.

More

Qwen3.8 Flash Next on every GPU and Mac · Every model that runs on the RTX 4070 Ti · Other models that fit 12 GB VRAM · Try other settings in the calculator

Guides: Best GPU for local LLMs by budget · Can I run gpt-oss, Llama 70B or Qwen3? · Mac vs PC for local LLMs

Data: Hugging Face Hub (Qwen/Qwen3.8-Flash-Next), NVIDIA specs. Model data last checked 2026-11-17; refreshed hourly.