mistralai/Magistral-Small-2509-GGUF
24B • Updated • 2.75k
• 75
Note Modern, supports vision and thinking - Q4_K_M with 8K context or IQ4_XS for 16K
Text Generation
• 21B • Updated • 6.66M
• • 5.07k
Note Beast for some tasks - use high thinking
nunchaku-ai/nunchaku-flux.1-krea-dev
Text-to-Image
• Updated • 23.1k
• 121
Note Realistic text to image
nunchaku-ai/nunchaku-flux.1-kontext-dev
Image-to-Image
• Updated • 4.1k
• 66
Note Great image editing
unsloth/Qwen3-30B-A3B-Instruct-2507-GGUF
31B • Updated • 550k
• 318
Note Very fast and reliable, good speed even with offloading as its an MOE
moondream/moondream3-preview
Image-Text-to-Text
• 9B • Updated • 131k
• 684
Note goated for Vision
Text-to-Speech
• Updated • 1.83M
• • 1.79k
Note Great TTS , supports voice cloning
google/gemma-3-27b-it-qat-q4_0-gguf
Image-Text-to-Text
• 27B • Updated • 405
• 403
Note All round great . Doesn't fit fully so it will be slow, use it for one off tasks
noctrex/Qwen3-Coder-REAP-25B-A3B-MXFP4_MOE-GGUF
Text Generation
• 25B • Updated • 341
• 15