Models that run on 256 GB

1161 models with published weights that fit in 256 GB — a Mac Studio with 256. Counted at four-bit quantisation, weights only, leaving the machine about a third of its memory and a gigabyte for context: room for roughly 274 billion parameters. A mixture of experts is counted in full, because it is held in full even though only a few experts compute. 1,161 in all, a hundred to a page; this is page 4 of 12.

A Mac Studio at the top of its configuration, or a serious machine. Almost everything openly published fits, including the large mixtures of experts. At this point the constraint is no longer whether the model loads but whether it generates fast enough to be worth waiting for.

• Kimi-VL-A3B-Thinking-2506Moonshot AI16.4B≈10.7 GB at 4-bit
• Wan2.1-I2V-14B-480PWan-AI16.4B≈10.7 GB at 4-bit
• Wan2.1-I2V-14B-720PWan-AI16.4B≈10.7 GB at 4-bit
• LLaDA2.0-UniInclusionAI16.3B≈10.6 GB at 4-bit
• Wan2.2-S2V-14BWan-AI16.3B≈10.6 GB at 4-bit
• Ling-mini-2.0InclusionAI16.3B≈10.6 GB at 4-bit
• Ring-mini-2.0InclusionAI16.3B≈10.6 GB at 4-bit
• Instella-MoE-16B-A3B-Thinkamd16.0B≈10.4 GB at 4-bit
• deepseek-moe-16b-basedeepseek-ai16.0B≈10.4 GB at 4-bit
• deepseek-moe-16b-chatdeepseek-ai16.0B≈10.4 GB at 4-bit
• Moonlight-16B-A3B-InstructMoonshot AI16.0B≈10.4 GB at 4-bit
• starcoder2-15bbigcode16.0B≈10.4 GB at 4-bit1 also selling it hosted
• starcoderbigcode15.8B≈10.3 GB at 4-bit
• DeepSeek-Coder-V2-Lite-InstructDeepSeek15.7B≈10.2 GB at 4-bit1 also selling it hosted
• DeepSeek-V2-Litedeepseek-ai15.7B≈10.2 GB at 4-bit
• starchat-alphaHugging Face H415.5B≈10.1 GB at 4-bit
• Apriel-1.6-15b-ThinkerServiceNow-AI15.0B≈9.8 GB at 4-bit
• Phi-4-reasoning-vision-15BMicrosoft15.0B≈9.8 GB at 4-bit
• WizardCoder-15B-V1.0WizardLM Team15.0B≈9.8 GB at 4-bit
• starcoder2-15b-instruct-v0.1bigcode15.0B≈9.8 GB at 4-bit1 also selling it hosted
• Apriel-1.5-15b-ThinkerServiceNow-AI14.9B≈9.7 GB at 4-bit
• DeepCoder-14B-PreviewAgentica14.8B≈9.6 GB at 4-bit
• Qwen2.5-14B-Instruct-1MAlibaba14.8B≈9.6 GB at 4-bit
• Qwen2.5-Coder-14B-InstructAlibaba14.8B≈9.6 GB at 4-bit1 also selling it hosted
• Strand-Rust-Coder-14B-v1Fortytwo14.8B≈9.6 GB at 4-bit
• SuperNova-MediusArcee AI14.8B≈9.6 GB at 4-bit
• Qwen3 14BAlibaba14.8B≈9.6 GB at 4-bit8 also selling it hosted
• BAGEL-7B-MoTByteDance Seed14.7B≈9.5 GB at 4-bit
• Phi-4Microsoft14.7B≈9.5 GB at 4-bit6 also selling it hosted
• Qwen1.5-MoE-A2.7BAlibaba14.3B≈9.3 GB at 4-bit
• 14BCausalLM14.0B≈9.1 GB at 4-bit
• ChatTS-14Bbytedance-research14.0B≈9.1 GB at 4-bit
• DeepSeek R1 Distill QWEN 14BDeepSeek14.0B≈9.1 GB at 4-bit5 also selling it hosted
• F2LLM-v2-14Bcodefuse-ai14.0B≈9.1 GB at 4-bit
• Fathom-R1-14BFractalAIResearch14.0B≈9.1 GB at 4-bit
• Nemotron-Labs-Diffusion-14BNVIDIA14.0B≈9.1 GB at 4-bit
• Qwen2.5-14BAlibaba14.0B≈9.1 GB at 4-bit1 also selling it hosted
• Velvet-14BAlmawave14.0B≈9.1 GB at 4-bit
• WAN2.2-14B-Rapid-AllInOnePhr00t14.0B≈9.1 GB at 4-bit
• Wan-Dancer-14BWan-AI14.0B≈9.1 GB at 4-bit
• Wan2.1-T2V-14BWan-AI14.0B≈9.1 GB at 4-bit1 also selling it hosted
• rwkv-4-pile-14bBlinkDL14.0B≈9.1 GB at 4-bit
• miniGCausalLM14.0B≈9.1 GB at 4-bit
• Phi-3-medium-128k-instructMicrosoft14.0B≈9.1 GB at 4-bit1 also selling it hosted
• translategemma-12b-itGoogle13.2B≈8.6 GB at 4-bit
• NexusRaven-V2-13BNexusflow13.0B≈8.5 GB at 4-bit
• Llama-2-13b-hfMeta Llama13.0B≈8.5 GB at 4-bit
• Baichuan2-13B-ChatBaichuan Intelligent Technology13.0B≈8.5 GB at 4-bit
• CodeLlama-13b-Instruct-hfCode Llama13.0B≈8.5 GB at 4-bit
• LLaVA-13b-delta-v0liuhaotian13.0B≈8.5 GB at 4-bit
• Llama-2-13bMeta Llama13.0B≈8.5 GB at 4-bit
• Llama-2-13b-chatMeta Llama13.0B≈8.5 GB at 4-bit1 also selling it hosted
• Llama-2-13b-chat-hfMeta13.0B≈8.5 GB at 4-bit1 also selling it hosted
• Llama2-13B-TiefighterKoboldAI13.0B≈8.5 GB at 4-bit1 also selling it hosted
• Llama2-Chinese-13b-ChatFlagAlpha13.0B≈8.5 GB at 4-bit
• MythoMax 13BGryphe13.0B≈8.5 GB at 4-bit4 also selling it hosted
• Nous Hermes Llama2 13B13.0B≈8.5 GB at 4-bit2 also selling it hosted
• OpenOrca-Platypus2-13BOpenOrca13.0B≈8.5 GB at 4-bit
• Orca-2-13bMicrosoft13.0B≈8.5 GB at 4-bit
• ReMM SLERP 13BUndi9513.0B≈8.5 GB at 4-bit1 also selling it hosted
• WhiteRabbitNeo-13B-v1WhiteRabbitNeo13.0B≈8.5 GB at 4-bit
• Wizard-Vicuna-13B-UncensoredQuixi AI13.0B≈8.5 GB at 4-bit
• Wizard-Vicuna-13B-Uncensored-HFTheBloke13.0B≈8.5 GB at 4-bit
• WizardLM-13B-UncensoredQuixi AI13.0B≈8.5 GB at 4-bit
• WizardLM-13B-V1.2WizardLM Team13.0B≈8.5 GB at 4-bit
• Ziya-LLaMA-13B-v1Fengshenbang-LM13.0B≈8.5 GB at 4-bit
• chronos-hermes-13b-v2Austism13.0B≈8.5 GB at 4-bit2 also selling it hosted
• jais-13binception4213.0B≈8.5 GB at 4-bit
• jais-13b-chatinception4213.0B≈8.5 GB at 4-bit1 also selling it hosted
• llama-13bhuggyllama13.0B≈8.5 GB at 4-bit
• mythalion-13bPygmalionAI13.0B≈8.5 GB at 4-bit
• open_llama_13bOpenLM Research13.0B≈8.5 GB at 4-bit
• prometheus-13b-v1.0prometheus-eval13.0B≈8.5 GB at 4-bit
• ruGPT-3.5-13Bai-forever13.0B≈8.5 GB at 4-bit
• stable-vicuna-13b-deltaCarperAI13.0B≈8.5 GB at 4-bit
• vicuna-13b-v1.5Large Model Systems Organization13.0B≈8.5 GB at 4-bit
• vicuna-13b-v1.5-16kLarge Model Systems Organization13.0B≈8.5 GB at 4-bit
• Krea-2-RawKrea12.8B≈8.3 GB at 4-bit
• Krea-2-TurboKrea12.8B≈8.3 GB at 4-bit
• Wayfarer-12BLatitude12.2B≈8.0 GB at 4-bit
• Mistral NemoMistral AI12.2B≈8.0 GB at 4-bit10 also selling it hosted
• Gemma 3 12BGoogle12.2B≈7.9 GB at 4-bit9 also selling it hosted
• Gemma-3-R1984-12BVIDraft12.2B≈7.9 GB at 4-bit
• Mellum2-12B-A2.5B-ThinkingJetBrains12.1B≈7.9 GB at 4-bit
• Llama Guard 4 12BMeta12.0B≈7.8 GB at 4-bit6 also selling it hosted
• Huihui-gemma-4-12B-coder-fable5-composer2.5-v1-abliteratedhuihui-ai12.0B≈7.8 GB at 4-bit
• KaLM-Embedding-Gemma3-12B-2511Tencent Hunyuan12.0B≈7.8 GB at 4-bit
• MN-12B-Celeste-V1.9nothingiisreal12.0B≈7.8 GB at 4-bit
• NVIDIA-Nemotron-Nano-12B-v2NVIDIA12.0B≈7.8 GB at 4-bit2 also selling it hosted
• NemoMix-Unleashed-12BMarinaraSpaghetti12.0B≈7.8 GB at 4-bit
• Pixtral 12B 2409Mistral AI12.0B≈7.8 GB at 4-bit3 also selling it hosted
• gemma-3-12b-it-qat-q4_0-unquantizedGoogle12.0B≈7.8 GB at 4-bit
• oasst-sft-1-pythia-12bOpenAssistant12.0B≈7.8 GB at 4-bit
• oasst-sft-4-pythia-12b-epoch-3.5OpenAssistant12.0B≈7.8 GB at 4-bit
• pythia-12bEleutherAI12.0B≈7.8 GB at 4-bit1 also selling it hosted
• Gemma-4-12BGoogle12.0B≈7.8 GB at 4-bit
• Gemma-4-12B-itGoogle12.0B≈7.8 GB at 4-bit
• Huihui-gemma-4-12B-it-abliteratedhuihui-ai12.0B≈7.8 GB at 4-bit
• FLUX.1-Canny-devBlack Forest Labs11.9B≈7.7 GB at 4-bit
• FLUX.1-Depth-devBlack Forest Labs11.9B≈7.7 GB at 4-bit