What you need to run Gemma 4 31B

About Gemma 4 31B

Gemma 4 31B is a general chat model from Google. One of the best models that fits a 24 GB GPU or 32 GB Mac. Running it locally needs about 21.5 GB of memory, and the download is 20 GB.

Smallest Mac that runs it

The least memory you need, by Apple chip:

• M1: does not fit
• M1 Pro: does not fit
• M1 Max: from 32 GB
• M1 Ultra: from 64 GB
• M2: does not fit
• M2 Pro: does not fit
• M2 Max: from 32 GB
• M2 Ultra: from 64 GB
• M3: does not fit
• M3 Pro: does not fit
• M3 Max: from 36 GB
• M3 Ultra: from 96 GB
• M4: does not fit
• M4 Pro: from 24 GB
• M4 Max: from 36 GB
• M5: does not fit
• M5 Pro: from 24 GB
• M5 Max: from 36 GB

Graphics cards

Runs well on: RTX 3090, RTX 4090, RTX 5090, Radeon RX 7900 XTX. Runs slowly with help from system RAM on: RTX 3060 12GB, RTX 3070, RTX 3080 10GB, RTX 4060, RTX 4060 Ti 16GB, RTX 4070 / 4070 Super, RTX 4070 Ti Super, RTX 4080 / 4080 Super, RTX 5060, RTX 5060 Ti 16GB, RTX 5070, RTX 5070 Ti, RTX 5080, Radeon RX 7800 XT, Radeon RX 9070 XT.

Hardware to run it

As an Amazon Associate I earn from qualifying purchases.

• Smallest Mac: M5 Pro with 24 GB. See it on Amazon
• Smallest graphics card: RTX 3090 (24 GB). See it on Amazon
• Storage: this model is a 20 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

• gpt-oss 20B: needs about 15.2 GB, a lighter choice. OpenAI's open model. Smart for its size and fast because only part of it runs per word.
• Gemma 4 26B A4B: needs about 19.4 GB, a lighter choice. Mixture of experts: big-model quality at small-model speed.
• Qwen 3.5 35B A3B: needs about 23.6 GB, a heavier choice. Very fast for its size. Great general assistant.
• Llama 3.3 70B: needs about 45.6 GB, a heavier choice. Older but solid large model. Needs serious memory.

Questions