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

Can Nex N2.6 run on the RTX 5070?

Not fully in VRAM. Even at 2-bit with 8K context, Nex N2.6 needs about 118 GB, more than the 11.4 GB usable on the RTX 5070. It can still run with offload (slower): Q2_K with part of the model in system RAM, about 6.7–13 tokens/sec (estimate). Details and command.

Runs with offload (slower)

Runs with offload 2-bit (Q2_K) at 8K context: all attention layers plus the experts of 2 of 43 MoE layers on the RTX 5070 (about 8.9 GB), and the experts of the other 41 layers in system RAM (about 108 GB). Needs a PC with at least 128 GB of system RAM. Estimated speed: 6.7–13 tokens/sec (slow but usable), versus a bandwidth ceiling of ~94 tok/s if it fit fully in VRAM.

llama-server -hf bartowski/Nex-N2.6-GGUF:Q2_K -c 8192 -ngl 999 --n-cpu-moe 43

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) •118 GB No118 GB No118 GB No119 GB No—
3-bit (Q3_K_M) •145 GB No145 GB No145 GB No147 GB No—
4-bit (Q4_K_M) •188 GB No188 GB No188 GB No189 GB No—
5-bit (Q5_K_M) •228 GB No228 GB No228 GB No229 GB No—
6-bit (Q6_K) •262 GB No263 GB No263 GB No264 GB No—
8-bit (Q8_0) •319 GB No319 GB No319 GB No320 GB No—
16-bit (F16/BF16)599 GB No599 GB No600 GB No601 GB No—

• real GGUF file size from bartowski/Nex-N2.6-GGUF. Speed is a bandwidth ceiling (672 GB/s), not a benchmark. Methodology.

Hardware that does fit it

H200Mac M2 Ultra 192GBMac M3 Ultra 256GBMac M3 Ultra 512GB

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

Nex N2.6 on every GPU and Mac · Every model that runs on the RTX 5070 · 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 (nex-agi/Nex-N2.6), NVIDIA specs. Model data last checked 2026-11-17; refreshed hourly.