What you need to run gpt-oss 120B

About gpt-oss 120B

gpt-oss 120B is a general chat model from OpenAI. OpenAI's big open model. Fast for its size on high-memory machines. Running it locally needs about 68.8 GB of memory, and the download is 65 GB. It is a mixture of experts model: only about 3.3 GB of its weights are read for each word, so it generates much faster than a dense model of the same size, but all 65 GB still has to fit in memory.

Smallest Mac that runs it

The least memory you need, by Apple chip:

• M1: does not fit
• M1 Pro: does not fit
• M1 Max: does not fit
• M1 Ultra: from 128 GB
• M2: does not fit
• M2 Pro: does not fit
• M2 Max: from 96 GB
• M2 Ultra: from 128 GB
• M3: does not fit
• M3 Pro: does not fit
• M3 Max: from 96 GB
• M3 Ultra: from 96 GB
• M4: does not fit
• M4 Pro: does not fit
• M4 Max: from 128 GB
• M5: does not fit
• M5 Pro: does not fit
• M5 Max: from 128 GB

Graphics cards

No single desktop graphics card runs it well.

Hardware to run it

As an Amazon Associate I earn from qualifying purchases.

• Smallest Mac: M3 Ultra with 96 GB. See it on Amazon
• No single desktop graphics card runs it well.
• Storage: this model is a 65 GB download, and most people end up keeping several. A fast external SSD keeps your main drive free. See SSDs on Amazon

Lighter and heavier alternatives

• Qwen 3.5 35B A3B: needs about 23.6 GB, a lighter choice. Very fast for its size. Great general assistant.
• Llama 3.3 70B: needs about 45.6 GB, a lighter choice. Older but solid large model. Needs serious memory.
• Qwen 3.5 122B A10B: needs about 85.5 GB, a heavier choice. Near frontier quality at home, if you have 96 GB or more.
• DeepSeek V3 671B: needs about 424.7 GB, a heavier choice. Frontier class. Needs a 512 GB Mac Studio or a server.

Questions