Skip to content

Local AI models

Can my computer run gpt-oss 120B?

OpenAI's larger open model, for computers with a lot of memory: careful reasoning, tools, and coding. gpt-oss 120B needs about 66.7 GB of memory. It runs comfortably on Macs with 128 GB or more.

gpt-oss 120B on common computers

Graded with Toskar 1.14.0's own estimate: the model file plus working memory at an 8K context.

ComputerFitWhat to expect
Mac, 8 GBToo bigNeeds more memory than this computer has.
Mac, 16 GBToo bigNeeds more memory than this computer has.
Mac, 24 GBToo bigNeeds more memory than this computer has.
Mac, 32 GBToo bigNeeds more memory than this computer has.
Mac, 48 GBToo bigNeeds more memory than this computer has.
Mac, 64 GBTight fitRuns, with little memory to spare. Close other big apps.
Mac, 128 GBExcellent fitPlenty of memory to spare.
PC, 8 GB, no graphics cardToo bigNeeds more memory than this computer has.
PC, 16 GB, no graphics cardToo bigNeeds more memory than this computer has.
PC, 32 GB, no graphics cardToo bigNeeds more memory than this computer has.
PC, 16 GB + 8 GB graphics cardToo bigNeeds more memory than this computer has.
PC, 32 GB + 12 GB graphics cardToo bigNeeds more memory than this computer has.
PC, 32 GB + 16 GB graphics cardToo bigNeeds more memory than this computer has.
PC, 64 GB + 24 GB graphics cardTight fitRuns, with little memory to spare. Close other big apps. Partial GPU offload expected.

About gpt-oss 120B

Memory needed
≈ 66.7 GB
Download size
59.0 GB
Parameters
117B
Quantization
MXFP4
Longest conversation
128K tokens

Good at

  • Chat and writing
  • Coding
  • Step-by-step reasoning
  • Uses tools, such as web search and your files

Languages it handles well

English

Similar models to compare

  • Qwen 3.6 35B-A3B≈ 21.8 GB

    A large mixture of experts that runs fast for its size: each word uses 3B of its 35B parameters. Sees pictures too.

  • Qwen 2.5 32B≈ 21.5 GB

    Large high-quality model for capable machines.

  • Qwen 3.8 27B≈ 20.4 GB

    The most capable everyday assistant, for computers with plenty of memory. Sees pictures too.

  • Muse Glimmer 30B≈ 17.1 GB

    Meta's large model that sees pictures, codes, and reasons well; for 32 GB Macs or 24 GB graphics cards.

Questions

How much memory does gpt-oss 120B need?
About 66.7 GB with Toskar's default 8K context: the 59.0 GB model file plus room for the conversation. After it has run once, Toskar measures what it really uses on your computer.
Can gpt-oss 120B run on a Mac with 8 GB?
No. It needs about 66.7 GB, more than a Mac with 8 GB has. A smaller model, or a paired computer with more memory, can answer instead.
Can gpt-oss 120B run without a graphics card?
It needs more memory than most PCs without a graphics card have. No. It needs about 66.7 GB, more than a PC with 16 GB of RAM and no graphics card has. A smaller model, or a paired computer with more memory, can answer instead.
Is gpt-oss 120B free to use?
Yes. Toskar Core is free and open source, and it downloads gpt-oss 120B for you at no cost. Everything runs on your computer, so there's no per-question bill.

Every model

Run gpt-oss 120B with Toskar

Toskar Core is free and open source for Mac, Windows, and Linux. It checks what your computer can run, downloads the model for you, and keeps every question on your computer.