unsloth/Qwen3.5-0.8B-GGUF

Image-Text-to-Text • 0.8B • Updated Mar 2 • 318k • 247

Note Qwen3.5 0.8B in Outmake. Best at running on almost any Mac, even ones with little memory to spare.

unsloth/Qwen3.5-2B-GGUF

Image-Text-to-Text • 2B • Updated Mar 2 • 446k • 169

Note Qwen3.5 2B in Outmake. Best at fast everyday chat on lighter Macs, with picture understanding built in.

unsloth/Qwen3.5-4B-GGUF

Image-Text-to-Text • 4B • Updated Mar 2 • 1.39M • 535

Note Qwen3.5 4B in Outmake. Best at everyday chat and simple coding help while staying light on memory.

unsloth/Qwen3.5-9B-GGUF

Image-Text-to-Text • 9B • Updated Mar 2 • 1.65M • 1.06k

Note Qwen3.5 9B in Outmake. Best all-round pick for a 16GB Mac: writing, coding help, and pictures.

unsloth/gemma-4-E2B-it-GGUF

Image-Text-to-Text • 5B • Updated Jul 17 • 572k • 351

Note Gemma 4 E2B in Outmake. Best tiny option for pictures and quick chat on a low-memory Mac.

unsloth/gemma-4-E4B-it-GGUF

Image-Text-to-Text • 8B • Updated Jul 17 • 610k • 678

Note Gemma 4 E4B in Outmake. Best everyday pick for a 16GB Mac: chat, writing, and pictures.

unsloth/gemma-4-26B-A4B-it-GGUF

Image-Text-to-Text • 25B • Updated Jul 17 • 1.67M • 1.27k

Note Gemma 4 26B in Outmake. Best mix of speed and quality for a 24-36GB Mac, with pictures.

unsloth/gemma-4-31B-it-GGUF

Image-Text-to-Text • 31B • Updated Jul 17 • 337k • 667

Note Gemma 4 31B in Outmake. Best Gemma choice for a 48-64GB Mac wanting top everyday quality.

unsloth/Qwen3.8-27B-GGUF

27B • Updated Aug 20 • 5.99M • 5.51k

Note Qwen3.8 27B in Outmake. Best pick for a 24-36GB Mac wanting strong all-round reasoning.

unsloth/Devstral-Small-2-24B-Instruct-2512-GGUF

24B • Updated Dec 15, 2025 • 99.4k • 194

Note Devstral Small 2 24B in Outmake. Best pick for coding help on a 24-36GB Mac.

unsloth/GLM-4.7-Flash-GGUF

Text Generation • 30B • Updated Feb 12 • 245k • 744

Note GLM 4.7 Flash in Outmake. Best pick for fast, capable chat and agent tasks on a 24-36GB Mac.

unsloth/gpt-oss-20b-GGUF

Text Generation • 21B • Updated Dec 19, 2025 • 589k • 899

Note GPT-OSS 20B in Outmake. Best pick for a 24GB+ Mac wanting OpenAI-style reasoning and tool use.

unsloth/gpt-oss-120b-GGUF

Text Generation • 117B • Updated Aug 25, 2025 • 73.6k • 301

Note GPT-OSS 120B in Outmake. Best pick for a 96GB+ Mac wanting OpenAI's strongest open reasoning model.

unsloth/Qwen3.6-35B-A3B-GGUF

Image-Text-to-Text • 35B • Updated Apr 20 • 1.47M • 1.77k

Note Qwen3.6 35B in Outmake. Best pick for a 48-64GB Mac wanting fast, high-quality everyday answers.

unsloth/Qwen3.5-397B-A17B-GGUF

Image-Text-to-Text • 396B • Updated Mar 7 • 21.5k • 254

Note Qwen3.5 397B in Outmake. Best flagship pick for a 192GB+ Mac wanting the strongest local answers, in a size for every big Mac.

unsloth/Qwen3.5-122B-A10B-GGUF

Image-Text-to-Text • 122B • Updated Mar 5 • 67k • 302

Note Qwen3.5 122B in Outmake. Best pick for a 96-128GB Mac wanting reasoning and picture understanding without the 397B flagship's size.

unsloth/Qwen3.8-Flash-Next-GGUF

Image-Text-to-Text • 177B • Updated 29 days ago • 1.18M • 1.32k

Note Qwen3.8 Flash Next in Outmake. Best pick for a 192-256GB Mac wanting Qwen's newest preview design, with pictures and a very long memory of the conversation.

unsloth/DeepSeek-V4-Flash-0731-GGUF

284B • Updated Aug 6 • 151k • 780

Note DeepSeek V4 Flash in Outmake. Best pick for a 192-256GB Mac wanting DeepSeek's fast coding and computer-use model, with a dial for how hard it thinks.

unsloth/MiniMax-M2.7-GGUF

Text Generation • 229B • Updated Apr 14 • 5.01k • 221

Note MiniMax M2.7 in Outmake. Best pick for a 192-256GB Mac wanting MiniMax's strongest agent and coding model, for personal or non-commercial use.

unsloth/GLM-5.3-GGUF

Text Generation • 754B • Updated Aug 29 • 354k • 115

Note GLM 5.3 in Outmake. Best flagship pick for a 512GB Mac wanting a top coding and agent model.

unsloth/Kimi-K2.6-GGUF

Image-Text-to-Text • 1T • Updated Apr 22 • 4.66k • 176

Note Kimi K2.6 in Outmake. Best pick for a 512GB Mac wanting the biggest model in this catalog, with picture understanding built in.