Update README.md

README.md

@@ -130,10 +130,6 @@ widget:
 
 ## Model description
 
-<h3 id="please-donate-buzz-for-flux-lora-training-!-9x2jxaes8"><span style="color:rgb(250, 176, 5)">Please Donate Buzz for FLUX Lora Training !</span></h3><p></p><h3 id="flux-v.2-(olsen-in-2017)-:-bnh357hrg"><strong><u>FLUX v.2 (Olsen in 2017) :</u></strong></h3><p>Trained on FLUX.1 [dev] with 80 photos of<span style="color:rgb(250, 176, 5)"> </span><strong><span style="color:rgb(250, 176, 5)">Elizabeth Olsen in 2017 with detailed GPT-4 captions</span></strong>. <strong><span style="color:rgb(34, 139, 230)">Tested on FLUX </span></strong><a target="_blank" rel="ugc" href="http://1.dev"><strong><span style="color:rgb(34, 139, 230)">1.dev</span></strong></a><strong><span style="color:rgb(34, 139, 230)"> (full) and FLUX fp8 and FLUX nf4 !</span><span style="color:#e64980"> No keywords needed.</span></strong> Use around strength <strong>0.8-1.0</strong>. Distilled CFG 3.5 and CFG 1.0 (without negative prompt). Clipskip 1. Can be used for example as follows:</p><p></p><p><strong><span style="color:rgb(130, 201, 30)">Positive</span></strong><span style="color:rgb(130, 201, 30)"> : {Artstyle, Character and scene description in usual FLUX fashion},  &lt;lora:eliolsen_2017_local_164_merger_20v1_8v2_34v2_03_03_04:1&gt;</span></p><p></p><h3 id="flux-v.1-(olsen-in-2012)-:-xcd7lsflf"><strong><u>FLUX v.1 (Olsen in 2012) :</u></strong></h3><p>Trained on FLUX.1 [dev] with 70 photos of<span style="color:rgb(250, 176, 5)"> </span><strong><span style="color:rgb(250, 176, 5)">Elizabeth Olsen in 2012 with detailed GPT-4 captions</span></strong>. <strong><span style="color:rgb(34, 139, 230)">Tested on FLUX </span></strong><a target="_blank" rel="ugc" href="http://1.dev"><strong><span style="color:rgb(34, 139, 230)">1.dev</span></strong></a><strong><span style="color:rgb(34, 139, 230)"> (full) and FLUX fp8 and FLUX nf4 !</span></strong> Use around strength <strong>0.8-1.0</strong>. Distilled CFG 3.5 and CFG 1.0 (without negative prompt). Clipskip 1. Can be used for example as follows:</p><p></p><p><strong><span style="color:rgb(130, 201, 30)">Positive</span></strong><span style="color:rgb(130, 201, 30)"> : {Artstyle, Character and scene description in usual FLUX fashion}, &lt;lora:Elizabeth_Olsen_2012_FLUX_epoch_46:1&gt;</span></p><p></p><h3 id="v8.0-2012:-bd0l22f1n"><strong><u>v8.0 2012:</u></strong></h3><p>Last versions probably. I wasn't satisfied with the likeness of the other models so I restricted all images to around 2012. Trained on Juggernaut X with 220 photos of<span style="color:rgb(250, 176, 5)"> </span><strong><span style="color:rgb(250, 176, 5)">Elizabeth Olsen in 2012</span></strong>. <strong><span style="color:rgb(34, 139, 230)">Tested on Juggernaut X, Juggernaut v7, RealismEngine 2, RealVisXL3 and AlbedoBase 2.0!</span></strong> <strong><span style="color:rgb(250, 82, 82)">Use with keyword : "</span><em><span style="color:rgb(250, 82, 82)">elxolsn</span></em><span style="color:rgb(250, 82, 82)">"</span> </strong>. Use around strength <strong>1.0-1.1</strong>. CFG 5.0-7.0. Clipskip 1. 10-40 steps. Can be used for example as follows:</p><p></p><p><strong><span style="color:rgb(130, 201, 30)">Positive</span></strong><span style="color:rgb(130, 201, 30)"> : {Artstyle}, {Character and scene description}, elxolsn, &lt;lora:eliolsen_2012_juggerX_xl_1_wocap-elxolsn-000120:1.05&gt;</span></p><p><strong><span style="color:rgb(230, 73, 128)">Negative</span></strong><span style="color:rgb(230, 73, 128)"> : ugly, deformed, airbrushed, photoshop, rendered, (multiple people), child</span></p><h3 id="-8o33a9ita"></h3><h3 id="v7.0-:-1nplymsen"><strong><u>v7.0 :</u></strong></h3><p>Trained on Juggernaut X with 347 photos of<span style="color:rgb(250, 176, 5)"> </span><strong><span style="color:rgb(250, 176, 5)">Elizabeth Olsen</span></strong>. <strong><span style="color:rgb(34, 139, 230)">Tested on Juggernaut X, Juggernaut v7, RealismEngine 2, RealVisXL3!</span></strong> <strong><span style="color:rgb(250, 82, 82)">Use with keyword : "</span><em><span style="color:rgb(250, 82, 82)">elixolsen</span></em><span style="color:rgb(250, 82, 82)">"</span> </strong>. Use around strength <strong>0.85-1.0</strong>. CFG 5.0-7.0. Clipskip 1. Can be used for example as follows:</p><p></p><p><strong><span style="color:rgb(130, 201, 30)">Positive</span></strong><span style="color:rgb(130, 201, 30)"> : {Artstyle}, {Character and scene description}, elixolsen, &lt;lora:eliolsen_juggerX_xl_8_standard_wocap_merger_lastmodel_120_04_06-elixolsen:0.95&gt;</span></p><p><strong><span style="color:rgb(230, 73, 128)">Negative</span></strong><span style="color:rgb(230, 73, 128)"> : ugly, deformed, airbrushed, photoshop, rendered, (multiple people), child</span></p><p></p><h3 id="v6.0-juggernaut-x-:-deuxhoe9r"><strong><u>v6.0 Juggernaut X :</u></strong></h3><p>Trained for Juggernaut X with 320 photos of Elizabeth Olsen. <strong>Use with Juggernaut X</strong>, Juggernaut v7 or <strong>RealismEngine 2</strong>. <strong>Works best with Juggernaut X</strong>. <strong><span style="color:rgb(250, 82, 82)">Use with keyword : "elixolsen"</span></strong>. Can be used for example as follows:</p><p></p><p><strong><span style="color:rgb(64, 192, 87)">Positive</span></strong><span style="color:rgb(64, 192, 87)"> : {Artstyle}, {Character and scene description}, elixolsen, &lt;lora:eliolsen_juggerX_v7_standard_wocap_merger_27_74_98_02_03_05-elixolsen:1&gt;</span></p><p><strong><span style="color:rgb(190, 75, 219)">Negative</span></strong><span style="color:rgb(190, 75, 219)"> : ugly, deformed, airbrushed, photoshop, rendered, (multiple people), child</span></p><p></p><h3 id="v5.0-160mb-:-98ykbaya0"><strong><u><span style="color:rgb(193, 194, 197)">v5.0 160mb :</span></u></strong></h3><p><span style="color:rgb(193, 194, 197)">Trained on 150 even more consistent (2015-2017) photos of </span><strong><span style="color:rgb(193, 194, 197)">Elizabeth Olsen</span></strong><span style="color:rgb(193, 194, 197)"> with a smaller network dimension (32 instead of 64) to allow for smaller model size by half. </span><strong><span style="color:rgb(250, 82, 82)">Use with keyword : "elixolsen"</span></strong><span style="color:rgb(250, 82, 82)">!</span></p><p></p><h3 id="v4.0:-872w5mfrl"><strong><u><span style="color:rgb(193, 194, 197)">v4.0:</span></u></strong></h3><p><span style="color:rgb(193, 194, 197)">Trained on </span><strong><span style="color:rgb(193, 194, 197)">347 high quality photos</span></strong><span style="color:rgb(193, 194, 197)"> of </span><strong><span style="color:rgb(193, 194, 197)">Elizabeth Olsen</span></strong><span style="color:rgb(193, 194, 197)"> between </span><strong><span style="color:rgb(193, 194, 197)">2015-2019</span></strong><span style="color:rgb(193, 194, 197)"> to ensure better facial consistency. Slight likeness improvement.</span></p><p></p><h3 id="v3.0:-kih3tzu48"><strong><u>v3.0:</u></strong></h3><p><span style="color:rgb(193, 194, 197)">Lora trained on 199 images of Elizabeth Olsen and captioned by GPT-4 Vision. Lora strength should be between </span><strong>0.8-1.2</strong><span style="color:rgb(193, 194, 197)">.</span></p><p></p><h3 id="v2.0:-5ewkepfda"><strong><u>v2.0:</u></strong></h3><p><span style="color:rgb(193, 194, 197)">Lora retrained on 100 images of Elizabeth Olsen with SDXL 1.0 base and an improved Lora architecture to increase flexibility and likeness. The recommended strength on </span><strong>ComfyUi </strong><span style="color:rgb(193, 194, 197)">is </span><strong>0.9-1.1</strong><span style="color:rgb(193, 194, 197)">.</span></p><p></p><h3 id="v1.0:-tkhegqvx9"><strong><u>v1.0:</u></strong></h3><p>Lora trained on 100 images of Elizabeth Olsen with SDXL 1.0 base. The recommended strength on <strong>ComfyUi </strong>is <strong>0.9-1.1</strong>.</p>
-
-
-
 ## Download model
 
 Weights for this model are available in Safetensors format.