Skip to content
Run LLM Locally

gpt-oss-120b on the RTX 6000 Ada

gpt-oss-120b at MXFP4 needs about 63.2 GB, more than the 48 GB of the RTX 6000 Ada: part of the model runs from system RAM, which slows generation.

ShareFacebookXWhatsAppLinkedInEmail

gpt-oss-120b

MXFP4 · 8K context

CPU/GPU offload

The model needs about 63.2 GB but only 46.6 GB of VRAM is usable: 16.6 GB must be loaded into system RAM. It runs, 74% on the GPU, but more slowly.

Your hardware
RTX 6000 Ada 48 GB · 64 GB RAM
Estimated memory
~63.2 GB estimate
VRAM (usable memory)
46.6 GB
GPU only
No
CPU/GPU offload
Required (16.6 GB)
Generation speed
21–41 tok/s estimate
Largest context in fast memory
None
Confidence: low

Based on: memory bandwidth (960 GB/s) and model size · system RAM bandwidth (~51 GB/s)

Memory breakdown

~63.2 GB (61.6 GB–64.8 GB) needed · estimate

16.6 GB must be offloaded to system RAM

  • Model weights59 GB
  • KV cache0.3 GB
  • Runtime overhead0.9 GB
  • Safety margin3 GB
Assumptions: context 8K tokens · KV cache f16 · llama.cpp (CUDA) · flash attention on · weights = real file size · sliding-window layers keep 128 tokens

EstimateData updated 24 days agoSource: Run LLM Locally engine, Hugging Face config.json + model card

What makes this estimate less reliable

  • no benchmark for this hardware and model yet
  • split CPU/GPU runs vary a lot with the CPU and PCIe link
  • system RAM speed unknown (dual-channel DDR4-3200 assumed)
  • mixture-of-experts: speed depends on expert routing

Other quantizations on the RTX 6000 Ada

gpt-oss-120b on RTX 6000 Ada + 64 GB RAM · 8K context
gpt-oss-120b MXFP48K~63.2 GB · 21–41 tok/s estimateCPU/GPU offload

Alternatives

GPUs that run gpt-oss-120b MXFP4 entirely in VRAM

Other models that fit the RTX 6000 Ada

Hardware that runs gpt-oss-120b MXFP4 entirely in memory

Prices and links follow the market you choose.

Radeon PRO W7900

AMD

ExcellentRuns gpt-oss-120b MXFP4 entirely in memory

Price not available

Availability unknown

View on amazon.com

Affiliate link

NVIDIA DGX Spark (GB10)

Unified memory PC

ExcellentRuns gpt-oss-120b MXFP4 entirely in memory

Price not available

Availability unknown

View on amazon.com

Affiliate link

No current offer imported for this market: the link opens a marketplace search.

Frequently asked questions

Can the RTX 6000 Ada run gpt-oss-120b?
The model needs about 63.2 GB but only 46.6 GB of VRAM is usable: 16.6 GB must be loaded into system RAM. It runs, 74% on the GPU, but more slowly.
Which quantization should I use on the RTX 6000 Ada?
None fits entirely in VRAM at 8K context; the smallest, MXFP4, still needs offload.
What is the largest context on the RTX 6000 Ada?
At MXFP4, the model does not fit entirely in VRAM even at a short context.
How fast is gpt-oss-120b (RTX 6000 Ada)?
We estimate 21–41 tok/s (estimate, confidence low). Based on: memory bandwidth (960 GB/s) and model size, system RAM bandwidth (~51 GB/s).