Agentic models that run on 96 GB
12 models with published weights that fit in 96 GB — a MacBook Pro with 96. Counted at four-bit quantisation, weights only, leaving the machine about a third of its memory and a gigabyte for context: room for roughly 102 billion parameters. A mixture of experts is counted in full, because it is held in full even though only a few experts compute.
An Apple configuration for people who intend to run models rather than occasionally try one. Comfortably holds the seventy-billion class with a very long context, or a mixture-of-experts model whose active parameters are few but whose weights are all resident.
- Qwen2.5 VL 72B InstructAlibaba73.4B≈47.7 GB at 4-bit7 also selling it hosted
- Qwen2.5 72B InstructAlibaba72.7B≈47.3 GB at 4-bit12 also selling it hosted
- Qwen2-72B-InstructAlibaba72.0B≈46.8 GB at 4-bit3 also selling it hosted
- Llama 3.3 70B InstructMeta70.6B≈45.9 GB at 4-bit29 also selling it hosted
- Llama-2-70b-chat-hfMeta70.0B≈45.5 GB at 4-bit1 also selling it hosted
- Llama-xLAM-2-70b-fc-rSalesforce70.0B≈45.5 GB at 4-bit1 also selling it hosted
- Meta-Llama-3.1-70B-InstructMeta70.0B≈45.5 GB at 4-bit7 also selling it hosted
- Holo3-35B-A3BH Company35.0B≈22.8 GB at 4-bit1 also selling it hosted
- QwQ-32B-PreviewAlibaba32.0B≈20.8 GB at 4-bit2 also selling it hosted
- xLAM-2-32b-fc-rSalesforce32.0B≈20.8 GB at 4-bit1 also selling it hosted
- Qwen3.5-27BAlibaba27.8B≈18.1 GB at 4-bit8 also selling it hosted
- gpt2-xlOpenAI community1.6B≈1.0 GB at 4-bit