Viggle Turbo V3 ships as full Qwen-Image fine-tune with GGUF and INT8 quants
linoy_tsaban · x · 2026-10-01
Viggle released V3 of its Qwen-Image-2.1-viggle-turbo model, and boosters stress it is a real full fine-tuned transformer, not a LoRA.
- Built on a 4-step DMD distillation setup with a step-400 EMA LoRA (r64) and scheduler config
- Ships merged single-file GGUF quant (6-step Q4KM) plus INT8/FP8 variants in a 105 GB repo
- Includes a ComfyUI workflow for the merged single-file transformers
- Model card documents the gap to the base model; tagline: "High quality can be fast."
- Licensed under qwen-research
More from Multimodal
- Runway x Claude Opus 5.5 video puts the whole coffee world in a cup, frame by frame — CurieuxExplorer · 2026-10-01
- Runway MCP Lands in ChatGPT App Directory: Generate Video From Any Agent — tlakomy · 2026-10-01
- ComfyUI test: Qwen Image 2.1 beats FLUX.2 Krea Turbo at character sheets, ~3x faster — cgpixel23 · 2026-10-01
- AI video tool NoSpoon shutting down to public, last demo made for $20 in 10 minutes — Kyrannio · 2026-10-01
- Midjourney partial tone-reversal prompt turns morning photos into glowing darkroom stills — tisch_eins · 2026-10-01
- Aiden Bai shares a prompt that generates a 30-second game trailer with original music and sound effects — aidenybai · 2026-10-01