Rent a GPU

Can the Apple M4 Pro (24GB) run gpt-oss 20B?

Apple M4 Pro (24GB) × gpt-oss 20B8K context
YesRuns fully on the GPU
Memory at Q5_K_M15.7 GB of 16.1 GB
Weights 14.9 GBKV cache 0.2 GBRuntime 0.6 GB
QuantQ5_K_MVery good
Speed~64tokens per second
Max context23Ktokens that fit at Q5_K_M
  1. FP16
  2. Q8_0
  3. Q6_K
  4. Q5_K_M
  5. Q4_K_M
  6. Q3_K_M
  7. Q2_K

Yes. The Apple M4 Pro (24GB) can run gpt-oss 20B fully in unified memory at Q5_K_M (very good), using about 15.7 GB of the 16.1 GB available, at roughly 64 tokens/second.

Try other context lengths

Run it with Ollama

With Ollama installed, one command downloads and starts gpt-oss 20B:

ollama run gpt-oss:20b

Ollama's default download is Q4_K_M (about 13.5 GB here). You have room for Q5_K_M, which is very good: pick that GGUF in LM Studio or from Hugging Face.

New to this? Step-by-step setup for the Apple M4 Pro (24GB)

Run it in the cloud

No Apple M4 Pro (24GB)? Run gpt-oss 20B in the cloud.

Rent a cloud GPU by the hour. No upfront cost, stop any time.

Rent a GPU on RunPodRent a GPU on Vast.ai

Memory needed by quantization

At 8,192 tokens of context. The black line marks the 16.1 GB the Apple M4 Pro (24GB) can use. macOS lets the GPU use about 16.1 GB of the 24 GB unified memory by default.

QuantQualityMemoryFits?Speed
FP16Full quality42.6 GBToo big—
Q8_0Near-lossless23 GBToo big—
Q6_KExcellent17.9 GBToo big—
Q5_K_MVery good15.7 GBFits~64 tok/s
Q4_K_MGood, most popular13.5 GBFits~75 tok/s
Q3_K_MNoticeably degraded11 GBFits~93 tok/s
Q2_KHeavily degraded8.6 GBFits~121 tok/s

Measured speed from real Apple M4 Pro (24GB) owners

Tokens per second that people measured on their own Apple M4 Pro (24GB), next to our estimate.

No one has measured gpt-oss 20B on the Apple M4 Pro (24GB) yet. The speeds above are estimates. Run the benchmark and yours becomes the first real number on this page.

Have an Apple M4 Pro (24GB)? Add your real speed

One command times gpt-oss 20B in your own Ollama, shows you exactly what it would send, and asks before sending anything. No sign-up, no personal data.

macOS and Linux

curl -fsSL https://gpufit.dev/bench.sh | sh -s -- gpt-oss:20b

Windows (PowerShell)

$env:GPUFIT_MODEL="gpt-oss:20b"; irm https://gpufit.dev/bench.ps1 | iex

How the benchmark works

GPUs that run gpt-oss 20B comfortably

Models that run well on the Apple M4 Pro (24GB)

Frequently asked questions

Can Apple M4 Pro (24GB) run gpt-oss 20B?

Yes. The Apple M4 Pro (24GB) can run gpt-oss 20B fully in unified memory at Q5_K_M (very good), using about 15.7 GB of the 16.1 GB available, at roughly 64 tokens/second.

How much VRAM does gpt-oss 20B need?

About 13.5 GB at Q4_K_M with 8,192 tokens of context, and about 42.6 GB at full FP16 precision.

How fast is gpt-oss 20B on the Apple M4 Pro (24GB)?

Roughly 64 tokens per second at Q5_K_M, estimated from the Apple M4 Pro (24GB)'s 273 GB/s memory bandwidth.