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

Can Llama 4 Scout 17B 16E Instruct run on the Arc B580?

Not fully in VRAM. Even at 2-bit with 8K context, Llama 4 Scout 17B 16E Instruct needs about 39.1 GB, more than the 11.4 GB usable on the Arc B580. It can still run with offload (slower): Q4_K_M with part of the model in system RAM, about 5.6–10 tokens/sec (estimate). Details and command.

Runs with offload (slower)

Runs with offload 4-bit (Q4_K_M) at 8K context: all attention layers plus the experts of 1 of 48 MoE layers on the Arc B580 (about 9.4 GB), and the experts of the other 47 layers in system RAM (about 53.0 GB). Needs a PC with at least 64 GB of system RAM. Estimated speed: 5.6–10 tokens/sec (slow but usable), versus a bandwidth ceiling of ~42 tok/s if it fit fully in VRAM.

llama-server -hf unsloth/Llama-4-Scout-17B-16E-Instruct-GGUF:Q4_K_M -c 8192 -ngl 999 --n-cpu-moe 47

Estimate, not a benchmark. The command downloads the GGUF from Hugging Face. 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, raise --n-cpu-moe by 1-2; if VRAM is left over, lower 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) •38.4 GB No39.1 GB No43.6 GB No61.6 GB No—
3-bit (Q3_K_M) •49.7 GB No50.5 GB No55.0 GB No73.0 GB No—
4-bit (Q4_K_M) •62.4 GB No63.2 GB No67.7 GB No85.7 GB No—
5-bit (Q5_K_M) •72.8 GB No73.6 GB No78.1 GB No96.1 GB No—
6-bit (Q6_K) •83.9 GB No84.6 GB No89.1 GB No107 GB No—
8-bit (Q8_0) •108 GB No109 GB No113 GB No131 GB No—
16-bit (F16/BF16) •202 GB No203 GB No208 GB No226 GB No—

• real GGUF file size from unsloth/Llama-4-Scout-17B-16E-Instruct-GGUF. Speed is a bandwidth ceiling (456 GB/s), not a benchmark. Methodology.

Hardware that does fit it

RTX A6000RTX 6000 AdaL40SMac M4 Pro 64GBMac M1 Max 64GBMac M4 Max 64GB

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

Llama 4 Scout 17B 16E Instruct on every GPU and Mac · Every model that runs on the Arc B580 · 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 (meta-llama/Llama-4-Scout-17B-16E-Instruct), Intel specs. Model data last checked 2026-11-17; refreshed hourly.