Source: https://docs.sogni.ai/sogni-create/studio/models/model-types/ # Model Types Sogni Studio supports several generations of image models, plus dedicated video and audio model families. They differ in architecture, prompt understanding, generation speed, and where they can run. ## [#](https://docs.sogni.ai/sogni-create/studio/models/model-types/#krea-2-turbo)Krea 2 Turbo [Krea 2 Turbo](https://www.sogni.ai/models/krea-2-turbo) is Sogni's newest open-source Krea-family image model. It uses a distilled few-step workflow for fast text-to-image and image-to-image, handles natural-language prompts, supports quoted in-image text, and renders up to 2560px on the Sogni Supernet. Its Sogni model id is `krea2_turbo_fp8_scaled`. ## [#](https://docs.sogni.ai/sogni-create/studio/models/model-types/#flux-family)Flux family Created by Black Forest Labs, the Flux suite is the current state of the art for prompt adherence, scene complexity, and style range. - **Flux.1 \[schnell\]** — fast distilled Flux, 1–4 steps. Good default for iteration. - **Chroma v.46 \[flash\] / v.48 \[detail calibrated\]** — stylized variants of Flux with distinctive color grading. Flux models run on the Sogni Supernet. ## [#](https://docs.sogni.ai/sogni-create/studio/models/model-types/#sd3)SD3 Stable Diffusion 3 uses a **triple text encoder** (CLIP + OpenCLIP + T5) and a **multi-modal diffusion transformer (MMDiT)** denoiser. The T5 encoder gives SD3 stronger text comprehension and natural-language prompt handling than earlier SD generations. ## [#](https://docs.sogni.ai/sogni-create/studio/models/model-types/#sdxl)SDXL Stable Diffusion XL produces 1024×1024 native output with stronger prompt adherence and better text rendering than SD 1.x/2.x. The XL models are the heaviest of the SD lineage but produce more photorealistic outputs with greater detail. Enabling upscaling pushes results to 2048×2048. ## [#](https://docs.sogni.ai/sogni-create/studio/models/model-types/#sdxl-turbo-and-lightning)SDXL Turbo and Lightning SDXL Turbo and SDXL Lightning are distilled SDXL variants designed for single-step or few-step generation (1–4 steps instead of 25–50). Quality is close to full SDXL at a fraction of the inference time — excellent for live iteration. ## [#](https://docs.sogni.ai/sogni-create/studio/models/model-types/#standard-sd-15-2x)Standard SD (1.5 / 2.x) Variants of Stable Diffusion 1.5, 2.0, and 2.1 — the foundation the open-source community trained hundreds of styles on. Don't be fooled by the version number: these models are still extremely capable and remain the lightest, fastest option for on-device generation. Most support ControlNet. ## [#](https://docs.sogni.ai/sogni-create/studio/models/model-types/#lcm-latent-consistency-models)LCM (Latent Consistency Models) LCM models are distilled versions of SD or SDXL that produce images in 2–8 steps instead of 25–50. Use the **LCM scheduler** when running them. Best with: - 2–8 inference steps - Guidance scale 1–2 - A Guide Image or ControlNet for structure Because LCM models converge so quickly, small changes in step count or guidance scale have outsized effects on output. Use a low-strength Guide Image (15–50) + a re-roll to upgrade an earlier render. ## [#](https://docs.sogni.ai/sogni-create/studio/models/model-types/#z-image)Z-Image Z-Image is a newer fast-inference image family. The `z_image_turbo_bf16` variant produces good results in a handful of steps and is one of Studio's best lightweight defaults for quick iteration. ## [#](https://docs.sogni.ai/sogni-create/studio/models/model-types/#qwen-image-edit)Qwen Image Edit Qwen Image Edit 2511 and Qwen Image Edit Plus are vision-language editing models — they take a source image plus a prompt and produce an edited variant that preserves identity, composition, and unaffected regions. They power Studio's Generative Filters and prompt-based edits in the in-app Chat. ## [#](https://docs.sogni.ai/sogni-create/studio/models/model-types/#video-models)Video models - **LTX-2.3** — image-to-video, optimized for motion fidelity. - **LTX-2.3 Dev** — text-to-video, latest generation, longer coherent shots. - **Wan 2.2** — first-party Sogni video model. - **[ByteDance Seedance](https://www.sogni.ai/models/seedance-2-5)** — Seedance 2.5 is the latest premium hosted tier and renders up to 1080p; full Seedance 2.0 remains available for 4K, while Seedance 2.0 Mini is the lower-cost 720p iteration path. - **[Alibaba HappyHorse 1.1](https://www.sogni.ai/models/happyhorse-1-1)** — premium hosted video for text prompts, one first-frame image, or 1-9 image references, with native synchronized audio. Video runs on the Supernet. See [Creating Videos](https://docs.sogni.ai/sogni-create/studio/creating-videos/) for which model fits which workflow. ## [#](https://docs.sogni.ai/sogni-create/studio/models/model-types/#audio-models)Audio models **ACE-Step** (turbo and SFT variants) generates music — full songs with AI-written lyrics, or instrumentals — with BPM, key, scale, and time-signature controls. See [Creating Audio](https://docs.sogni.ai/sogni-create/studio/creating-audio/). ## [#](https://docs.sogni.ai/sogni-create/studio/models/model-types/#on-device-quantizations-6-bit-45-bit-4-bit)On-device quantizations: \[6-bit\], \[4.5-bit\], \[4-bit\] Quantized model variants are about 3× lighter than their full-precision counterparts, with minor quality differences. They run on Macs with less unified memory and on older Apple Silicon. Use them when: - Your machine has less than 16 GB unified memory - You want to keep more models on disk without ballooning storage - You're chaining many models in a session and want to keep memory tight Full-precision versions are still the default on machines with the headroom. Quantized versions are macOS-only. * * * ▶️ **Tutorial video:** [Sogni AI Model Explorer: A World of Creative Styles](https://youtu.be/qLa6bbiThow?feature=shared) **Need help? Ask in our Discord!** Join the [Sogni Discord](https://discord.com/invite/2JjzA2zrrc) ✨