merge chroma and flux srpo model, I use lora <a target="_blank" rel="ugc" href="https://civarchive.com/models/1987788/flux-qwen-image-turbo">https://civarchive.com/models/1987788/flux-qwen-image-turbo</a>, which allows me to generate images in 8 steps minimum, with simple euler but other samplers are possible like euler A sgm uniform. you have to add the flux encoder texts as well as vae. I use the model with forge sd.

chroma-merge-srpo

merge chroma and flux srpo model, I use lora <a target="_blank" rel="ugc" href="https://civitai.com/models/1987788/flux-qwen-image-turbo">https://civitai.com/models/1987788/flux-qwen-image-turbo</a>, which allows me to generate images in 8 steps minimum, with simple euler but other samplers are possible like euler A sgm uniform. you have to add the flux encoder texts as well as vae. I use the model with forge sd.

chromaMergeSrpo_v10.safetensors

v1.0

this is a fusion model with svdq-int4_r32-flux.1-dev and svdquantInt4CreartUltimate_nunchaku030Version at 50/50 this model is not a nunchaku model, it will be necessary to add the encoder texts and usual vaeeuler simple and beta <span style="background-color:rgb(26, 27, 30);color:rgb(193, 194, 197);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">(DPM++ 2M SGM uniform seems fine too)10steps<a target="_blank" rel="ugc" href="https://civarchive.com/models/1545303?modelVersionId=1861654">https://civarchive.com/models/1545303?modelVersionId=1861654</a>

nunchaku-merge-dev

this is a fusion model with svdq-int4_r32-flux.1-dev and svdquantInt4CreartUltimate_nunchaku030Version at 50/50 this model is not a nunchaku model, it will be necessary to add the encoder texts and usual vaeeuler simple and beta10 steps

nunchakuMergeDev_v10.safetensors

FluxVeris combines real-time flow processing with a compact, high-performance architecture to produce faithful, reproducible outputs. A dedicated text encoder provides precise semantic conditioning (natural-language prompts → latent conditions), while the VAE compresses and reconstructs high-dimensional signals into a stable latent space that preserves realistic detail. The pipeline is optimized for controllable iterative refinement: sampling with 6–10 denoising steps covers typical speed/quality trade-offs, with 8 steps recommended as a practical compromise between fast generation and faithful realism.Key features<ul><li>Text encoder for accurate semantic conditioning and prompt-following.</li><li>VAE latent backbone to preserve fine detail while reducing computational cost.</li><li>Tunable sampling steps (6–10):<ul><li>6 steps — fastest, acceptable quality for exploratory/low-latency use.</li><li>8 steps — recommended sweet spot: good balance of quality and speed.</li><li>10 steps — higher fidelity for critical outputs where time is less constrained.</li></ul></li><li>Transparent, reproducible outputs designed for real-world fidelity and auditability.</li><li>Real-time friendly: engineered for low latency on modern inference hardware.</li></ul>

FluxVeris

FluxVeris combines real-time flow processing with a compact, high-performance architecture to produce faithful, reproducible outputs. A dedicated text encoder provides precise semantic conditioning (natural-language prompts → latent conditions), while the VAE compresses and reconstructs high-dimensional signals into a stable latent space that preserves realistic detail. The pipeline is optimized for controllable iterative refinement: sampling with 6–10 denoising steps covers typical speed/quality trade-offs, with 8 steps recommended as a practical compromise between fast generation and faithful realism.

fluxveris_.safetensors

<h3 id="learning">LEARNING</h3><h3 id="merge-model">Merge model</h3><h3 id="simple-euler-8-to-10-steps">Simple Euler 8 to 10 steps</h3><h3 id="text-encoder-and-vae-to-add">Text encoder and VAE to add</h3><h3 id="added-lora-lightning-for-10-step-generation.">added LORA Lightning for 10-step generation.</h3><a target="_blank" rel="ugc" href="https://civarchive.com/models/1987788?modelVersionId=2250142">https://civarchive.com/models/1987788?modelVersionId=2250142</a>

flux-qwensrpo

LEARNINGMerge modelSimple Euler 8 to 10 stepsText encoder and VAE to add

Files with SHA256: 84d7d9c1aedee88f3bbff2a752dacdf6cdc3262064339959bf744e52ea9b5ba3