OpenAI on Apple M3

Can I run gpt-oss 120B-A5.1B on an Mac Studio M3 Ultra 256GB?

Yes. gpt-oss 120B-A5.1B at Q4_K_M uses 67.1 GB of the 192.0 GB available on a Mac Studio M3 Ultra 256GB, and runs at about 107 tokens per second. You can push the context to 128K.

Runs comfortably

67.1 GB of 192.0 GB · 35%
0192 GB
Weights 66.0 GB
KV cache 0.6 GB
Runtime overhead 0.5 GB

gpt-oss 120B-A5.1B at Q4_K_M leaves 124.9 GB spare on a Mac Studio M3 Ultra 256GB. There is room to raise the context length or move up a quantisation level.

Generation107tok/s
Prompt processing2347tok/s
Max context128Ktokens
KV per 1K tokens0GB

Every quantisation of gpt-oss 120B-A5.1B on a Mac Studio M3 Ultra 256GB

Highlighted row is the highest quality that still fits at 8K context.

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 217.6 GB 218.6 GB Won't fit — 16.9
INT8 / W8A8 8.50 115.3 GB 116.4 GB Runs comfortably 128K 68.5
FP8 (E4M3) 8.00 108.8 GB 109.8 GB Runs comfortably 128K 72.0
Q6_K 6.56 89.2 GB 90.3 GB Runs comfortably 128K 84.8
Q5_K_M 5.67 77.1 GB 78.2 GB Runs comfortably 128K 95.3
Q5_K_S 5.52 75.1 GB 76.1 GB Runs comfortably 128K 97.3
Q4_K_M 4.85 66.0 GB 67.1 GB Runs comfortably 128K 107
Q4_K_S 4.58 62.4 GB 63.5 GB Runs comfortably 128K 112
Q4_0 4.55 62.0 GB 63.1 GB Runs comfortably 128K 113
AWQ 4-bit 4.25 59.4 GB 60.4 GB Runs comfortably 128K 116
GPTQ 4-bit 4.25 59.4 GB 60.4 GB Runs comfortably 128K 116
MXFP4 4.25 59.4 GB 60.4 GB Runs comfortably 128K 116
IQ4_XS 4.25 58.0 GB 59.0 GB Runs comfortably 128K 118
Q3_K_M 3.91 53.4 GB 54.4 GB Runs comfortably 128K 126
IQ3_M 3.70 50.6 GB 51.6 GB Runs comfortably 128K 131
IQ3_XXS 3.06 41.9 GB 43.0 GB Runs comfortably 128K 149
Q2_K 2.63 36.1 GB 37.2 GB Runs comfortably 128K 164
IQ2_XXS 2.06 28.5 GB 29.5 GB Runs comfortably 128K 190
IQ1_M 1.75 24.3 GB 25.4 GB Runs comfortably 128K 207

Also worth checking

gpt-oss 120B-A5.1B on every GPU
Full hardware tableEverything that runs on a Mac Studio M3 Ultra 256GB
Full model tableWhy the KV cache is the part that bites
Guide