Analog Photography LoRA for Z-Image Turbo

• Add the AnalogRedmond LoRA to Z-Image Turbo for a classic analog film photography look — grain, warm tones, and vintage compositions.
• Trigger it with analog + AnalogRedmAF in your prompt; the author recommends ComfyUI for best results.
• Licensed under Apache-2.0, free to use commercially and modify.

What AnalogRedmond does

AnalogRedmond is a style LoRA trained by ArtificialGuyBR on top of the Tongyi-MAI Z-Image Turbo base model. Per the Hugging Face model card, it "specializes in generating authentic analog photography with that classic film look," covering vintage portraits, nostalgic street photography, warm film aesthetics, and timeless photographic impressions.

The model card describes a high capacity for film grain textures and classic photography compositions. It is a style transfer LoRA rather than a character or object LoRA: it reshapes the output's photographic character instead of adding a specific subject.

Training and release background

The author credits GPU time donated by Redmond.AI for making the model possible. The weights are published in Safetensors format; the Civitai release is version v1.0 for Z-Image, an 81.15 MB file published in February 2026. Sibling versions of the style exist for other base models, including Qwen Image, Flux Klein 9B, and a v2.0 — this page covers the Z-Image Turbo variant.

Community reception has been strong: on Civitai the model page shows thousands of downloads (over 7,400 in the Hugging Face stats snapshot) with positive reviews and thousands of ratings, a notable adoption level for a niche style LoRA.

Trigger words

Use both trigger words together to activate the style reliably:

• AnalogRedmAF — the primary trained trigger, listed on Hugging Face as the instance prompt along with ANALOG.
• analog — the natural-language cue also documented in the model card.

The Civitai page lists AnalogRedmAF and Analog as the official trained words.

How to use it

ComfyUI (recommended by the author). Load the Safetensors file as a LoRA node attached to the Z-Image Turbo checkpoint, set your prompt including the trigger words, and generate. The model card does not state a specific LoRA strength, so start near the default weight and tune up or down to taste — higher strength pushes the film look harder at the cost of composition flexibility.

Diffusers. The Hugging Face repository is tagged as a diffusers text-to-image LoRA, so it can be loaded with DiffusionPipeline and LoraLoaderMixin against the Tongyi-MAI/Z-Image-Turbo base, then activated with the same trigger phrase.

Z-Image workflows elsewhere. Any interface that supports Z-Image Turbo LoRAs will accept this file; keep the trigger words in the prompt either way.

If you want the same analog look on a different base model, check the author's other Redmond-style variants (Qwen Image version, Flux Klein 9B version), or browse more Z-Image Turbo LoRAs in the portfolio. For prompt-writing help around film looks, see the analog film style prompt guide and the ComfyUI LoRA setup guide.

Common errors and fixes

Error | Cause | Fix | Source

|---|---|---|---

Style not appearing | Trigger words missing or misspelled | Always include analog and AnalogRedmAF in the prompt | HF model card Wrong base model mismatch | LoRA loaded on SDXL/FLUX instead of Z-Image Turbo | Load the Z-Image Turbo checkpoint and attach the LoRA there | Civitai page (base model: ZImageTurbo) Too-strong film effect | LoRA weight too high | Reduce the LoRA strength / weight node | Author recommendation in model card Blurry or soft output outside ComfyUI | Unsupported or non-standard Z-Image workflow | Use ComfyUI, the author's recommended tool | HF model card

FAQ

What base model does this LoRA need? Tongyi-MAI/Z-Image-Turbo. The Civitai listing marks the base model as ZImageTurbo; the Hugging Face metadata lists it both as the base model and the adapter target.

Can I use it commercially? Yes — the model is licensed under Apache License 2.0, which permits commercial use, modification, and redistribution.

Does it need a trigger word? Yes, it is not a zero-shot style LoRA. Use AnalogRedmAF plus analog.

Is there a version for other base models? Yes. The author publishes sibling variants for Qwen Image, Flux Klein 9B, and a v2.0, all sharing the AnalogRedmond style.

How big is the file? 81.15 MB in Safetensors format, per the Civitai v1.0 – Z-Image release.

Sources

• Hugging Face model card — ANALOG-REDMOND-ZIMAGETURBO: description, trigger words, base model, Apache-2.0 license, usage recommendation.
• Civitai — AnalogRedmond Analog Style Photography LoRA: version details, stats, reviews, trigger words, file size.