ChrisColeTech/qwen-image-edit-turbo-GGUF

Qwen-Image-Edit 2511 Turbo — GGUF build

20B instruction-driven image editing with the Lightning 8-step LoRA pre-merged. CFG-free; the transformer fits a single 24 GB card.

What this repo is: a requantized, split-layout build of the upstream model — weights only, not a retrain. The recommended settings below are the values these weights are actually run with day to day, not copied from the upstream card.

---

Samples

**prompt:** `a red apple on a rustic wooden table beside a window, soft daylight, visible wood grain, studio photograph, sharp focus` — 1024×1024, 8 steps, guidance 1.0, true_cfg 1.0, seed 7	**prompt:** `portrait of an older fisherman with a weathered face, natural window light, shallow depth of field, photorealistic` — 1024×1024, 8 steps, guidance 1.0, true_cfg 1.0, seed 99
**prompt:** `a rain-slicked Tokyo street at night, neon signs reflecting in puddles, cinematic, 35mm photograph` — 1024×1024, 8 steps, guidance 1.0, true_cfg 1.0, seed 1234	**edit instruction:** `place a small red wizard hat on top of the apple` — 8 steps, guidance 1.0, true_cfg 1.0, seed 5
**edit instruction:** `change the background to a snowy forest at dusk, keep the apple and table exactly as they are` — 8 steps, guidance 1.0, true_cfg 1.0, seed 6	**edit source** — the input both edits above were applied to. Compare the apple's speckled skin against the edited versions to judge how much surface detail an edit preserves.

---

img2img

This model does img2img as well as txt2img. The left image is the input; the right is what this model made of it.

**source prompt:** `a red apple on a rustic wooden table beside a window, soft daylight, visible wood grain, studio photograph, sharp focus` — 1024×1024, seed 7 — generated separately as the img2img input	**img2img prompt:** `the same scene rendered as a vivid oil painting, thick visible brushstrokes, rich saturated colour, painterly texture` — 1024×1024, **8 steps**, guidance 1.0, true_cfg 1.0, seed 7 — about 210 s on an A100 80GB

No strength is set, and none is needed. This is an instruction-edit model: it reads the prompt as an edit to apply rather than denoising the source by a fraction. A sweep at 0.3 / 0.5 / 0.7 / 0.9 produced four near-identical outputs — strength is inert here, which is why the preset omits it.

That is also why this model preserves the scene most faithfully of the family: the window, the table, the apple's position and its cast shadow all stay where they were while the surface becomes thick impasto brushwork. The others trade composition for style as strength rises; this one does not have to.

Recommended settings

Values this build is run with in practice. They are read out of a live config when this card is generated, so they cannot drift from what is actually used.

Parameter	Production value	Meaning
steps	8	Denoising steps
guidance	1.0	Guidance scale
true_cfg_scale	1.0	True-CFG scale (>1 enables the negative branch)

Supported modes: txt2img, img2img

Notes and gotchas

• true_cfg_scale is pinned at 1.0. The Lightning distillation is CFG-free: at 1.0 the negative branch is disabled and each step is one transformer forward, versus two above 1.0. That makes 1.0 unambiguously right for speed. Whether raising it would improve edit quality has not been measured on this build.
• Raising steps does not help. 8, 12 and 16 were rendered from the same source image, instruction and seed and are visually indistinguishable — the extra steps only cost time (~81s / ~122s / ~162s per edit on an A100). 8 is what the Lightning LoRA is distilled for.
• Edit texture loss was a quantisation problem, and Q6_K largely fixes it. On the previous Q4_K_M build, edits returned the subject with its surface detail smoothed and colours re-saturated — a hat edit turned the source apple's speckled, droplet-covered skin into flat glossy red. Step count was ruled out first (8, 12 and 16 are visually indistinguishable), leaving quantisation. Re-rendering every sample on Q6_K at the identical preset restored visible speckle: edge variance rose from 103.7 to 126.8 on the hat edit and 693.1 to 895.1 on the background swap, and the samples above are the Q6_K renders. Caveat on how tight that measurement is: the Q4 edits were made from an earlier source image that no longer exists, so the two sets were not driven from a byte-identical input — the effect is large and in the predicted direction, but it is not a perfectly controlled A/B.
• guidance is inert on this build (guidance_embeds=false on the transformer config) but is kept explicit for clarity.
• No width/height in the preset — this is deliberate. Omitting them lets the pipeline preserve the input image's aspect ratio.
• There is no strength parameter — this build has no partial-denoise branch.
• Fallback if you need the base build: transformer qwen-image-edit-2511-Q4_K_M.gguf with {steps: 40, guidance: 1.0, true_cfg_scale: 4.0}.

---

Layout

Components ship as separate files rather than one bundle: the transformer, the text encoder(s) and the VAE each load independently, with configs and tokenizers under config/. Any loader that accepts explicit per-component paths can consume this directly — point it at the files listed below.

A generic diffusers snippet is deliberately omitted. This is a repacked split layout with substituted encoders, so an upstream example will not load it unmodified, and an untested snippet would be worse than none.

---

Files

File	Size	Role
split/text_encoders/Qwen2.5-VL-7B-Instruct-q4_0.gguf	4.13 GB	text encoder
split/text_encoders/Qwen2.5-VL-7B-Instruct-mmproj-f16.gguf	1.26 GB	vision projector (mmproj)
split/qwen_image_vae.safetensors	242.05 MB	VAE
config/processor/tokenizer.json	10.89 MB	tokenizer / processor
config/tokenizer/tokenizer.json	6.71 MB	tokenizer / processor
config/tokenizer/vocab.json	3.23 MB	tokenizer / processor
config/processor/vocab.json	2.65 MB	tokenizer / processor
config/tokenizer/merges.txt	1.59 MB	tokenizer / processor
config/transformer/diffusion_pytorch_model.safetensors.index.json	194.23 KB	transformer (main weights)
config/text_encoder/model.safetensors.index.json	56.30 KB	text encoder
config/processor/tokenizer_config.json	4.62 KB	tokenizer / processor
config/tokenizer/tokenizer_config.json	4.58 KB	tokenizer / processor
config/text_encoder/config.json	3.07 KB	text encoder
config/tokenizer/config.json	1.34 KB	tokenizer / processor
config/processor/video_preprocessor_config.json	910 B	tokenizer / processor
config/processor/preprocessor_config.json	826 B	tokenizer / processor
config/vae/config.json	730 B	VAE
config/processor/special_tokens_map.json	613 B	tokenizer / processor
config/tokenizer/special_tokens_map.json	613 B	tokenizer / processor
config/processor/added_tokens.json	605 B	tokenizer / processor
config/tokenizer/added_tokens.json	605 B	tokenizer / processor
config/model_index.json	516 B	config
config/scheduler/scheduler_config.json	485 B	scheduler config
config/transformer/config.json	362 B	transformer (main weights)
config/tokenizer/preprocessor_config.json	350 B	tokenizer / processor
config/text_encoder/generation_config.json	244 B	text encoder

---

Provenance

• Upstream base model: Qwen/Qwen-Image-Edit-2511 (upstream license: apache-2.0)
• This build: requantized and relaid out into a split component layout. Weights are not retrained; only the format and directory layout differ.
• License: left as unknown in this repo's metadata. Refer to the upstream model's license for redistribution and commercial-use terms — several of these bases are non-commercial.