<h1 id="sophos-realism">Sophos Realism</h1>Long time lurker, first time poster. I recently set out to create an Illustrious-based model that blends realism, Danbooru prompt comprehension, and character comprehension. I think this model achieved that goal enough to merit seeing the light of day, so here it is.You need to use a VAE with this model; there is no VAE baked in. The default SDXL VAE works fine.<h2 id="version-notes">📝 Version Notes</h2>Consult the notes below for the differences between the posted versions.<h3 id="v1.6-ez-mode">v1.6 - EZ Mode</h3>This is just v1.5 with some helpful LoRAs merged in to make it easier to use. Use v1.5 instead of v1.6 if you want to have more control or don't like these LoRAs.<ul><li><a rel="ugc" href="https://civarchive.com/models/633300/dark-dramatic-chiaroscuro-lighting-slidersntcaixyz-notrigger">Dark</a> @ 0.3 strength. This enhances most lighting effects, IMO.</li><li><a rel="ugc" href="https://civarchive.com/models/1157318?modelVersionId=1388246">Photorealistic Skin</a> @ 0.25 strength. This helps with skin and I think improves realism overall.</li><li><a rel="ugc" href="https://civarchive.com/models/1461427/il-tool-native-high-resolution-super-wide-shot-better-highres-fix">Wide Shot</a> @ 0.5 strength. Thanks to this one, I recommend generating at a higher resolution like 1536 x 1536, 1344 x 1792, etc.</li></ul><h3 id="v1.5-refinement">v1.5 - Refinement</h3>Adjusted a little of this and a little of that. I'd say this version is a slight improvement over v1.4 in terms of polish and realism while maintaining good prompt adherence.<h3 id="v1.4-expressivev2">v1.4 - ExpressiveV2</h3>A tweak to the v1.3 formula that I think improved it overall.<h3 id="v1.3-expressive">v1.3 - Expressive</h3>This version achieves improved expressiveness over v1.2 while remaining a balanced checkpoint in terms of realism and prompt comprehension. It does character expressions especially well, but you might find v1.2 is better at some compositions. v1.1 remains the best checkpoint for "maximum realism" if that's your priority.I'd say use v1.3 if you like v1.2 but you need a little more creativity or range of character expressions.<h3 id="v1.2-balanced">v1.2 - Balanced</h3>This version sits somewhere between v1.0 and v1.1 in terms of realism and prompt comprehension. It seems to be about as flexible as v1.0 but with better overall realism.<h3 id="v1.1-realistic">v1.1 - Realistic</h3>Use this version if you're willing to sacrifice some prompt comprehension and LoRA compatibility for more realism. It's great if you just need character portraits, and it can do some more complicated compositions if you work at it, but its prompt comprehension isn't as good as the other checkpoints.<h3 id="v1.0-semi-realistic-more-anime">v1.0 - Semi-Realistic, More Anime</h3>This version should really be considered semi-realistic. It still looks very anime most of the time, but it can do almost everything an anime model can do, which can be fun. Use the recommended LoRAs below to nudge it more in a realistic direction.<h2 id="suggested-uses">Suggested Uses</h2>You can use this model more or less like you would use an Illustrious anime model, but the results look realistic. It understands danbooru tags and does a decent job of realistically rendering popular fictitious characters. It's pretty good at realism too, just generally speaking. If you want a more realistic edge to the finished photo, try using this model to get the generation started and then switch to your favorite SDXL realism model for refinement of those fine details.<h3 id="usage-tips">Usage Tips</h3><ul><li>Add Emphasis to Tags: You will have to add more emphasis than you might normally use to trigger the effect of certain danbooru tags for unrealistic things that you wouldn't normally see in real photos. In short, if it doesn't work at first, try upping the weight until it does (hopefully).</li><li>Boost Character Details: For characters, it helps to add some of their defining characteristics and also add more weight to the character name to guide the effect. For example, "aerith gainsborough" is okay, but "(aerith gainsborough:1.2), green eyes, single braid, pink ribbon in hair" is going to be even better. You don't really need to do that most of the time for the Illustrious anime models, but it helps with this one.</li><li>Use Concept/Chracter LoRAs: LoRAs are your friend. If a particular character or concept isn't well represented in the danbooru dataset, the effect of the trigger word might be too weak. Search Civit for a LoRA that can help. Odds are you'll find one... or dozens. Don't be afraid to apply those LoRAs at a lower weight. Sometimes a little dab will do ya.</li><li>👉 Use Recommended LoRAs: (This is only important for the v1.0 model.) See below! The v1.0 model thrives on two LoRAs in particular. The later models don't necessarily need them but I think they still benefit from them to achieve the best results.</li></ul><h3 id="nsfw-note">⚠️ NSFW Note ⚠️</h3>This model will do NSFW very nicely. It didn't lose any of that Illustrious potency, that's for sure. Please be responsible with what you do with it.<h2 id="recommended-prompts">Recommended Prompts</h2>Obviously tweak these for your exact needs. They aren't always going to be right for the effects you want, but they give generally good results.<h3 id="positive">Positive</h3><pre><code>(best quality:1.2) (photorealistic:1.2) (realistic:1.2) (raw photo:1.2) intricate details (detailed skin texture:1.2) (natural lighting:1.2) volumetric lighting (film grain:1.2) (depth of field:1.2)</code></pre><h3 id="negative">Negative</h3><pre><code>big eyes, worst quality, low quality, bad quality, text, logo, username, signature, colourless, impassive, big eyes, worst quality, low quality, bad quality, text, logo, username, signature, colourless, impassive, (text, logo, username, signature:1.2), photoshop, doll, (silicone:1.1), fake, filter, airbrush, 3d max, colourless, impassive, shaders, (anime eyes:1.2), (glossy skin:1.2), (plastic skin:1.2), cartoon proportions, illustration, digital art, simplified, abstract, (unrealistic:1.2), impressionistic, (lowres:1.2), (bad anatomy:1.2), (bad hands:1.2), (missing fingers:1.2), (worst quality:1.2), (low quality:1.2), (cartoon:1.1), (anime:1.1), drawing, sketch, (artificial:1.1)</code></pre>Note: You can prune quite a few of those tags out of the negative prompt and still get perfectly fine results. Play around with it. None of this is gospel.<h3 id="embeddings">Embeddings</h3>In my testing, this model works fine with popular positive/negative embeddings like <a target="_blank" rel="ugc" href="https://civarchive.com/models/1028231/realismillustriousnegativeembedding">Realism_Illustrious_Negative_Embedding</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/1028256?modelVersionId=1153237">Realism_Illustrious_Positive_Embedding</a>. Give them a try if that's your thing.<h2 id="loras">LoRAs</h2><h3 id="lora">LoRA</h3>The LoRAs are the secret sauce for the v1.0 version of this model. (No longer needed for v1.1.) This LoRA combination really juices the v1.0 model and makes it shine. They are a must for getting the results you see in the example images.<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/633300?modelVersionId=708019">dark (dramatic chiaroscuro lighting)</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/971952/stabilizer-ilnai">Stabilizer IL/NAI</a></li></ul>You can go as easy or hard on the dark LoRA as you want depending on your needs, but ~0.3 is a fine setting for general purpose use cases to enhance any image.The Stabilizer LoRA adds a little extra creativity and prompt adherence. Good values seem to be in the 0.1 - 0.4 range. If you go much higher than that, you'll get artifacts. I think ~0.2 is the sweet spot for general purpose usage.<pre><code>&lt;lora:illustriousXLv01_stabilizer_v1.185c:0.2&gt; &lt;lora:dark:0.3&gt;</code></pre><h3 id="other-lora-compatibility">Other LoRA Compatibility</h3>Generally speaking, any LoRA that works with the Illustrious anime models should work just fine on this model, although depending on the effect, it may not translate over nicely to the realism style. However, I've been pleasantly surprised with the results so far! Pretty much any Illustrious or SDXL LoRA, and even some Pony LoRAs, should work. Just remember that your mileage may vary.<h2 id="sampler-settings">Sampler Settings</h2><h3 id="samplers">Samplers</h3>Play around with your favorites, but here's what tends to work for me:<ul><li>DPM++ 2M SDE</li><li>DPM++ 3M SDE</li><li>ER SDE</li><li>Gradient Estimation (my favorite in reForge)</li><li>Anywhere between 25 and 60 steps works fine, with diminishing returns past ~40 steps.</li></ul><h3 id="cfg">CFG</h3>I use 3.5 CFG with Mahiro CFG in reForge. If you use regular CFG, try it in the 4 - 5 range, or use whatever works best for you.<h3 id="clip-skip">Clip Skip</h3>You can do either clip skip 1 or clip skip 2. The difference is very slight, but I think I lean towards preferring clip skip 2.<h3 id="scheduler">Scheduler</h3>I like the Phi scheduler if you have it available, otherwise you can't go wrong with SGM Uniform, Beta, or Karras. Ultimately use whatever works best for you.<h2 id="merge-details">Merge Details</h2>Credit goes where credit is due. This merge wouldn't have been possible without the efforts of the people who finetune the base models. Please join me in saying thank you to these fine creators. 🙏<h3 id="illustrious-models">Illustrious Models</h3>v1.0<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/1125067?modelVersionId=1973422">CyberIllustrious v5</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/928145/jedpointreal">JedpointReal v3</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/1277670?modelVersionId=1772645">JANKU v4</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/1346879?modelVersionId=1846026">MAGIC_ILL v2</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/376130?modelVersionId=1854228">Nova Anime XL v8</a></li></ul>v1.1<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/1125067?modelVersionId=1973422">CyberIllustrious v5</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/928145/jedpointreal">JedpointReal v3</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/1110783/ilustmix">iLustMix v7</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/1046064?modelVersionId=1575331">ILustREAL v5</a></li></ul>v1.2This model is a blend of v1.1 with <a target="_blank" rel="ugc" href="https://civarchive.com/models/1675671/hyphoria-real-illu">Hyphoria Real</a>, which is a great Illustrious realism model in its own right. You should definitely check it out if you like these models at all.v1.3This checkpoint is a blend of v1.2 with <a target="_blank" rel="ugc" href="https://civarchive.com/models/428826/damn-illustriouspony-realistic-model">DAMN! by coyotte</a>, which is a solid Illustrious realism model that you should totally check out.v1.4Merged v1.3 with a blend of other Illustrious realism models, some of which were already present in the blend.<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/928145/jedpointreal">JedpointReal v3</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/1675671/hyphoria-real-illu">Hyphoria Real v0.9</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/428826/damn-illustriouspony-realistic-model">DAMN! by coyotte v5</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/1820231?modelVersionId=2059829">MAGIC_ILL_PhotoReal v1</a></li></ul>v1.5I got lazy with this one and sat on it for a while, so I don't remember what all went into it. It wasn't much changed beyond v1.4. I think I merged some iLustREAL v5 back into it? I don't remember now, sorry.<h3 id="sdxl-models">SDXL Models</h3>v1.0 and v1.1 (present in other versions that merge v1.0 and v1.1)<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/194768?modelVersionId=1966530">Jib Mix Realistic XL v18</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/573152?modelVersionId=1569593">LUSTIFY! OLT</a></li></ul>

Sophos Realism

Illustrious-XL-v1.0.safetensors

sophosRealism_v10SemiRealistic.safetensors

v1.0 Semi-realistic

v1.6 - EZ Mode

v1.5 - Refinement

v1.4 - ExpressiveV2

v1.3 - Expressive

v1.2 Balanced

v1.1 - Priortized Realism

This is a quantized version of WAI-illustrious-SDXL, Built-in VAE and CLIP included for efficient inference.All models—Unet, Clip, and Vae—are included in a single ZIP file, complete with sample workflows and a brief guide.ComfyUI RecommendedBase Model: <a target="_blank" rel="ugc" href="https://civarchive.com/models/827184/wai-illustrious-sdxl">https://civarchive.com/models/827184/wai-illustrious-sdxl</a> All credits to the original creator: <a target="_blank" rel="ugc" href="https://civarchive.com/user/WAI0731">https://civarchive.com/user/WAI0731</a><hr />Support &amp; Custom Requests: <span style="background-color:rgb(26, 27, 30);color:rgb(141, 141, 143);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">Processing, testing, and optimizing these models require significant computational resources and time. If you require a specific Checkpoint or LoRA to be optimized for your hardware constraints, you may submit a formal request and support this ongoing work through the link below: <a target="_blank" rel="ugc" href="https://ko-fi.com/morikomorizz">https://ko-fi.com/morikomorizz</a>

WAI-illustrious-GGUF

waiIllustriousGGUF_vae.safetensors

WAI-V16-Repack

Workflow-GGUF

CLIP-L

CLIP-G

WAI-V15-Repack

This is a quantized version of Pony Diffusion V6 XL, Built-in VAE and CLIP included.The unet model only needs 2.8gb vram on KSampler / Processing Image.ComfyUI RecommendedBase Model: <a target="_blank" rel="ugc" href="https://civarchive.com/models/257749/pony-diffusion-v6-xl">https://civarchive.com/models/257749/pony-diffusion-v6-xl</a>All credits to the original creator: <a target="_blank" rel="ugc" href="https://civarchive.com/user/PurpleSmartAI">https://civarchive.com/user/PurpleSmartAI</a>Support &amp; Custom Requests: <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">Processing, testing, and optimizing these models require significant computational resources and time. If you require a specific Checkpoint or LoRA to be optimized for your hardware constraints, you may submit a formal request and support this ongoing work through the link below: <a target="_blank" rel="ugc" href="https://ko-fi.com/morikomorizz">https://ko-fi.com/morikomorizz</a>

Pony-Diffusion-GGUF

ponyDiffusionGGUF_vae.safetensors

Pony-Diffusion-V6-XL-GGUF

V6-Repack

Files with SHA256: 3cc3b7f04550109b453ff14dc0f91da4aed05f0b90bbc5bbb573c5eabf5161f0