<h3 id="please-donate-buzz-for-flux-lora-training-!-5so15u1an">Please Donate Buzz for FLUX Lora Training !</h3><h3 id="flux-v5.0-(2016-2018):-u6xz3lmsi">FLUX v5.0 (2016-2018):</h3>Trained on FLUX dev with 100 selected photos of Ana de Armas (2016-2018) with detailed GPT-4 captions. Tested on FLUX 1.D (full) and FLUX fp8 and FLUX nf4 ! Use around strength 0.9-1.2. No keywords needed! Distilled CFG around 1-4 and CFG 1.0 (without negative prompt). Clipskip 1. Can be used for example as follows:Positive : {Artstyle, Character and scene description in usual FLUX fashion},&lt;lora:Ana_de_Armas_v10_2016_2018_FLUX-000061:1&gt;<h3 id="flux-v4.0-:-jwnmugs28">FLUX v4.0 :</h3>Trained on FLUX dev with 90 selected photos of Ana de Armas with detailed GPT-4 captions. Tested on FLUX 1.D (full) and FLUX fp8 and FLUX nf4 ! Use around strength 0.9-1.2. No keywords needed! Distilled CFG around 1-4 and CFG 1.0 (without negative prompt). Clipskip 1. Can be used for example as follows:Positive : {Artstyle, Character and scene description in usual FLUX fashion},&lt;lora:Aa_de_Armas_FLUX_v5-merger_37_53_65_02_03_05:1.0&gt;<h3 id="flux-v3.0-:-v3vk9dfot">FLUX v3.0 :</h3>Improved likeness compared to v2.0.Trained on FLUX dev with 96 more carefully selected photos of Ana de Armas with detailed GPT-4 captions. Tested on FLUX 1.D (full) and FLUX fp8 and FLUX nf4 ! Use around strength 0.8-1.2. No keywords needed! Distilled CFG around 1-4 and CFG 1.0 (without negative prompt). Clipskip 1. Can be used for example as follows:Positive : {Artstyle, Character and scene description in usual FLUX fashion},&lt;lora:anarmas_local_v4_96img_merger_24_60_01_09:1&gt;<h3 id="flux-v2.0-:-tkqkgk8ps">FLUX v2.0 :</h3>Improved consitency and likeness.Trained on FLUX dev with 120 photos of Ana de Armas with detailed GPT-4 captions. Tested on FLUX 1.D (full) and FLUX fp8 and FLUX nf4 ! Use around strength 0.8-1.2. No keywords needed! Distilled CFG around 1-4 and CFG 1.0 (without negative prompt). Clipskip 1. Can be used for example as follows:Positive : {Artstyle, Character and scene description in usual FLUX fashion}, &lt;lora:anarmas_local_v2_large-merge_41v1_11v3_04_06_05_11v4_05:1&gt;<h3 id="flux-v1.0-:-7hoc8l9zb">FLUX v1.0 :</h3>Trained on FLUX dev with 80 photos of Ana de Armas with detailed GPT-4 captions. Tested on FLUX <a target="_blank" rel="ugc" href="http://1.dev">1.dev</a> (full) and FLUX fp8 and FLUX nf4 ! Use around strength 0.8-1.2. No keywords needed! Distilled CFG around 1-4 and CFG 1.0 (without negative prompt). Clipskip 1. Can be used for example as follows:Positive : {Artstyle, Character and scene description in usual FLUX fashion}, &lt;lora:Ana_de_Armas_FLUX_v1-000061:1&gt;<h3 id="sdxl-v4.0-juggernaut-x-:-oa1ph8wwa">SDXL v4.0 Juggernaut X :</h3>Trained for Juggernaut X with 250 photos of Ana de Armas. Use with Juggernaut X, Juggernaut v7 or RealismEngine 2. Works best with Juggernaut X. Use with keyword : "anxarmas". Can be used for example as follows:Positive : {Artstyle}, {Character and scene description}, anxarmas, &lt;lora:anarmas_juggerX_xl_3_standard-anxarmas-000096:1&gt;Negative : ugly, deformed, airbrushed, photoshop, rendered, (multiple people), child<h3 id="sdxl-v3.0-:-i2pwe7ug8">SDXL v3.0 :</h3>Trained on 250 more carefully selected images of Ana de Armas for 100 epochs with single repeat. The dataset captioning uses GPT-4 Vision. Recommended Lora strength is 0.8-1.2.<h3 id="sdxl-v2.1-:-bt3020zrv">SDXL v2.1 :</h3>Retrained on the same 420 images but with a better Lora model for more flexibility while attaining likeness. Keep strength between 0.9-1.1.<h3 id="sdxl-v2.0-:-noijub90c">SDXL v2.0 :</h3>Trained on 420 images of Ana de Armas for 100 epochs with single repeat and batch size 3. The dataset captioning is also done using a different method. Recommended ComfyUi Lora strength is 1.0-1.15.<h3 id="-6adr2rwkk"></h3><h3 id="sdxl-v1.0-:-4ydn9tqjq">SDXL v1.0 :</h3>Lora trained on 40 images of Ana de Armas on the SDXL 1.0 base model. The recommended Lora strength for ComfyUi is 0.9-1.2.

Ana de Armas (FLUX + SDXL)

Retrained on the same 420 images but with a better Lora model for more flexibility while attaining likeness. Keep strength between 0.9-1.1.

Files with SHA256: a7fd3e3b052e039c41d1b327c5c8c3b7873718f7cf3fb24f3d39b8df4b612d86