gpt-oss-120b hardware requirements

About the model

Parameters
116.8B total · 5.1B active (MoE)
Attention
Sliding-window attention
Max context
131,072 tokens
Released
2025-08-05

GGUF files

Tracked quantizations of gpt-oss-120b, smallest first
QuantFile sizeQualityPublished by
MXFP4baseline63.39 GBNear-losslessllama.cpp team · ggml-org

Memory needed by quant and context

Weights + f16 KV cache + compute buffer, in GB. Add about 0.6 GB if the GPU also drives your display on Windows.
Quant4k8k16k32k64k128k
MXFP464.164.364.665.466.969.9

Which GPUs can run gpt-oss-120b?

How to read the verdicts

Runs great
Fully on the GPU at 20 tok/s or more
Runs well
Fully on the GPU at 8–20 tok/s; MoE experts in system RAM at 20 tok/s or more; or at least 90% on the GPU at 8 tok/s or more
Runs slowly
2–8 tok/s; CPU-only; less than 90% on the GPU; or MoE experts in system RAM below 20 tok/s
Won't run
Does not fit, or under 2 tok/s

Scroll sideways to see every column.

54 GPUs, Macs and CPU setups at MXFP4 with 8k context, sorted by verdict and speed
HardwareVerdictSpeedQuantMemoryRuns asContextNotesPrice
Apple M2 Ultra · 128 GB
Runs great
79.7 tok/s8k estimate 75.6 tok/s (60.5–90.7, calibrated estimate ±20%)measured (1 run, 2k context)
MXFP464.3 / 89.6 GBUnified memoryup to 128k——
Apple M3 Ultra · 96 GB
Runs great
est. 77.4 tok/s61.9–92.8calibrated estimate ±20%
MXFP464.3 / 67.2 GBUnified memoryup to 64k——
Apple M5 Max (40-core GPU) · 128 GB
Runs great
est. 60.0 tok/s48.0–72.0calibrated estimate ±20%
MXFP464.3 / 89.6 GBUnified memoryup to 64k——
Apple M4 Max (40-core GPU) · 128 GB
Runs great
est. 53.4 tok/s42.7–64.0calibrated estimate ±20%
MXFP464.3 / 89.6 GBUnified memoryup to 64k——
Ryzen AI Max+ 395 (Strix Halo) · 128 GB
Runs great
50.2 tok/s8k estimate 44.2 tok/s (35.4–53.0, calibrated estimate ±20%)measured (1 run, 2k context)
MXFP464.3 / 89.6 GBUnified memoryup to 32k—$1,999 launch MSRP
NVIDIA DGX Spark · 128 GB
Runs great
35.0 tok/s8k estimate 35.6 tok/s (28.5–42.7, calibrated estimate ±20%)measured (1 run, 2k context)
MXFP464.3 / 89.6 GBUnified memoryup to 16k—$3,999 launch MSRP
DDR5-6000 dual-channel CPU · 96 GB
Runs slowly
est. 13.4 tok/s10.8–16.1calibrated estimate ±20%
MXFP463.7 GB RAMCPU onlyup to 128k
  • CPU inference — capped at Runs slowly
—
DDR5-5600 dual-channel CPU · 96 GB
Runs slowly
est. 12.6 tok/s10.0–15.1calibrated estimate ±20%
MXFP463.7 GB RAMCPU onlyup to 128k
  • CPU inference — capped at Runs slowly
—
Apple M4 · 32 GB
Won't runDetails
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 28.0 GB of memory is free (22.4 GB usable by the GPU)
—
Apple M4 Max (32-core GPU) · 48 GB
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 44.0 GB of memory is free (33.6 GB usable by the GPU)
—
Apple M4 Pro · 64 GB
Won't runDetails
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 60.0 GB of memory is free (44.8 GB usable by the GPU)
—
Apple M5 · 32 GB
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 28.0 GB of memory is free (22.4 GB usable by the GPU)
—
Apple M5 Max (32-core GPU) · 48 GB
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 44.0 GB of memory is free (33.6 GB usable by the GPU)
—
Apple M5 Pro · 64 GB
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 60.0 GB of memory is free (44.8 GB usable by the GPU)
—
Intel Arc B580
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 11.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$290 street (as of 2026-08-09)
DDR4-3200 dual-channel CPU · 64 GB
Won't run
—MXFP4needs 63.7 GB——
  • Needs about 63.7 GB; 60.0 GB of RAM is free
—
GeForce GTX 1080 Ti
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 10.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$699 launch MSRP
GeForce RTX 2080 Ti
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 10.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$999 launch MSRP
GeForce RTX 3050 8GB
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 7.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$249 launch MSRP
GeForce RTX 3060 12GB
Won't runDetails
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 11.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$250 street (as of 2026-08-09)
GeForce RTX 3070
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 7.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$499 launch MSRP
GeForce RTX 3080 10GB
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 9.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$699 launch MSRP
GeForce RTX 3080 12GB
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 11.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$799 launch MSRP
GeForce RTX 3090
Won't runDetails
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 23.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$1,050 street (as of 2026-08-09)
GeForce RTX 3090 Ti
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 23.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$1,999 launch MSRP
GeForce RTX 4060
Won't runDetails
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 7.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$299 launch MSRP
GeForce RTX 4060 Laptop
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 7.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
—
GeForce RTX 4060 Ti 16GB
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 15.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$499 launch MSRP
GeForce RTX 4060 Ti 8GB
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 7.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$399 launch MSRP
GeForce RTX 4070
Won't runDetails
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 11.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$599 launch MSRP
GeForce RTX 4070 Laptop
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 7.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
—
GeForce RTX 4070 Super
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 11.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$599 launch MSRP
GeForce RTX 4070 Ti Super
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 15.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$799 launch MSRP
GeForce RTX 4080
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 15.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$1,199 launch MSRP
GeForce RTX 4080 Laptop
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 11.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
—
GeForce RTX 4080 Super
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 15.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$999 launch MSRP
GeForce RTX 4090
Won't runDetails
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 23.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$2,755 street (as of 2026-08-09)
GeForce RTX 4090 Laptop
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 15.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
—
GeForce RTX 5060
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 7.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$339 street (as of 2026-08-09)
GeForce RTX 5060 Ti 16GB
Won't runDetails
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 15.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$569 street (as of 2026-08-09)
GeForce RTX 5060 Ti 8GB
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 7.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$429 street (as of 2026-08-09)
GeForce RTX 5070
Won't runDetails
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 11.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$629 street (as of 2026-08-09)
GeForce RTX 5070 Ti
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 15.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$949 street (as of 2026-08-09)
GeForce RTX 5070 Ti Laptop
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 11.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
—
GeForce RTX 5080
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 15.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$1,256 street (as of 2026-08-09)
GeForce RTX 5080 Laptop
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 15.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
—
GeForce RTX 5090
Won't runDetails
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 31.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$4,200 street (as of 2026-08-09)
GeForce RTX 5090 Laptop
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 23.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
—
Radeon RX 7800 XT
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 15.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$499 launch MSRP
Radeon RX 7900 XT
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 19.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$899 launch MSRP
Radeon RX 7900 XTX
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 23.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$999 launch MSRP
Radeon RX 9060 XT 16GB
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 15.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$449 street (as of 2026-08-09)
Radeon RX 9070
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 15.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$639 street (as of 2026-08-09)
Radeon RX 9070 XT
Won't run
—MXFP4needs 64.3 GB——
  • Needs about 64.3 GB; 15.4 GB of VRAM and 28.0 GB of RAM are free
  • With 96 GB RAM: Runs slowly
$689 street (as of 2026-08-09)

Reasoning models spend extra tokens thinking, so their speed thresholds are 1.5× stricter (30 / 12 / 3 tok/s).

Cheapest GPUs that run it

No GPU with a current street price reaches Runs great or Runs well at MXFP4. The table above lists Macs, unified-memory PCs and CPU setups.

Measured results for gpt-oss-120b

Public benchmarks we calibrate against. Their conditions (context, backend, flags) can differ from the estimates above.

Measured results for gpt-oss-120b
HardwareQuantBackendContextPrompt (tok/s)Generation (tok/s)FlagsSourceMeasured
Apple M2 UltraMXFP4llama.cpp2k1,245.079.7—github.com2025-08-15
Ryzen AI Max+ 395 (Strix Halo)MXFP4llama.cpp2k707.050.2ROCm7quozul.dev2025-10-27
NVIDIA DGX SparkMXFP4llama.cpp2k1,717.035.0—github.com2025-10-14
GeForce RTX 5090MXFP4llama.cpp4k—9.6--n-cpu-moe 21, DDR4 serverhardware-corner.net2026-08-09

Frequently asked questions

How much VRAM does gpt-oss-120b need?

At MXFP4 the weights are 63.39 GB; with 8k context the total is about 64.3 GB (KV cache 0.3 GB, compute buffer 0.6 GB).

What is the cheapest GPU that runs gpt-oss-120b well?

No GPU with a current street price reaches Runs great or Runs well at MXFP4. The best tracked option is Apple M2 Ultra · 128 GB: Runs great.

Can gpt-oss-120b run on an 8 GB, 16 GB or 24 GB GPU?

At MXFP4 with 8k context: GeForce RTX 4060 — Won't run; GeForce RTX 5060 Ti 16GB — Won't run; GeForce RTX 3090 — Won't run.

Speeds are estimates from memory bandwidth, calibrated against public benchmarks, and each one comes with an error band and a confidence label. Real results vary with drivers, backend, context length and thermals.