Model
Trends
.ai
Model
Trends
.ai
die besten Open-Source-KI-Modelle

Die bestplatzierten KI-Modelle, nach Kategorie und Zeitraum

Der ganze Katalog in Heat-Reihenfolge, die heißesten zuerst. Suche und filtere, um zu sehen, wo ein Modell unter seinesgleichen steht. It's over 9000.
Suchen
Nach Zeitraum filtern:
Nach Tag filtern:
NSFW
27,512 Modelle · Seite 1 von 331
  1. #1
    #1
    FLUX 2 / KLEIN
    •
    BASE
    FLUX.2 [dev] is Black Forest Labs' second-generation open-weights image model, released in November 2025. It is a much larger flow-matching transformer than FLUX.1 that uses the Mistral Small 3.1 vision-language model as its text encoder,…
    9832°
    Gold · Gesamt
  2. #2
    #2
    FLUX
    •
    BASE
    The fastest Flux DEV fp8 workflow, supports up to 5 loras. Built-in upscaler is a factor of 1.4, modify with /slot8
    9822°
    Silber · Gesamt
  3. #3
    #3
    QWEN
    •
    BASE
    Qwen-Image is the image generation foundation model of Alibaba's Qwen team, released as open weights in August 2025. It is a 20-billion-parameter multimodal diffusion transformer that reads prompts through the Qwen2.5-VL vision-language…
    9730°
    Bronze · Gesamt
  4. #4
    #4
    STABLE DIFFUSION XL 1.0
    •
    BASE
    Stable Diffusion XL version 1.0 is here, and the Refiner 1.0 is also installed. This is the official full release, but our software needs adjustment to fully support it. Only the /render command works and SDXL cannot be mixed with old…
    9658°
  5. AD
    Keine Installation nötig – KI-Modelle in der Cloud ausführen
    BitVector Prism ist der einfachste Einstieg: Modell wählen, Prompt eingeben und in einer übersichtlichen Web-App generieren, ohne etwas einzurichten. BitVector gibt es auch auf Discord und im Web (SpyGlass).
  6. #5
    #5
    STABLE DIFFUSION 1.5
    •
    BASE
    Stable Diffusion 1.5 by Stability / RunwayML, the OG model. Someone said it was weird that we didn't have this in the concepts collection, and they were correct, so here you go! Of course, this model is over a year old, but it's still…
    9546°
  7. #6
    #6
    STABLE DIFFUSION XL 1.0
    •
    BASE
    9457°
  8. #7
    #7
    STABLE DIFFUSION XL 1.0
    •
    LORA
    9300°
  9. #8
    #8
    9234°
  10. #9
    #9
    STABLE DIFFUSION XL 1.0
    •
    LORA
    The "hyper-xl" model is optimized to significantly increase generation speed.
    9231°
  11. #10
    #10
    9191°
  12. #11
    #11
    STABLE DIFFUSION 1.5
    •
    BASE
    waifu-diffusion by hakurei. A long running base model trained on crisp-looking anime. Compared to your other anime base model options, this one is fairly well-rounded, which unfortunately has been detrimental in making it a clear winner in…
    9139°
  13. #12
    #12
    STABLE DIFFUSION 1.5
    •
    BASE
    Dreamshaper 8 by Lykon is one of the most popular base models for illustrations and art. This update includes better support for character LoRAs, better photorealism, can still handle anime, and better NSFW
    8996°
  14. #13
    #13
    STABLE DIFFUSION XL 1.0
    •
    BASE
    This is the 1024 Aesthetic version of the general purpose Playground 2 concept, a rebuild of SDXL. Try it as a second opinion when you don't like what gives you. It can handle realism as well as paintings and ilustrations
    8928°
  15. #14
    #14
    STABLE DIFFUSION 1.5
    •
    BASE
    majicMIX realistic 麦橘写实 model by Merjic
    8926°
  16. #15
    #15
    FLUX
    •
    BASE
    8906°
  17. AD
    PirateDiffusion
    Keine Installation nötig – KI-Modelle in der Cloud ausführen
    PirateDiffusion gibt es nur auf Telegram und richtet sich an Profis: tausende Modelle, LoRAs und Workflows per Chat-Befehl, mit unbegrenzter Generierung zum Festpreis.
  18. #16
    #16
    8903°
  19. #17
    #17
    8901°
  20. #18
    #18
    STABLE DIFFUSION XL 1.0
    •
    BASE
    This is the old version of Illustrious by request, version beta .01. The new full version 1 is also available but works best as a workflow, or partially with render when using euler_a
    8885°
  21. #19
    #19
    STABLE DIFFUSION 1.5
    •
    BASE
    CounterfeitV3.0 by rqdwdw. The 3.0 version of Counterfeit has much brighter colors and more focus on the subject, whereas the previous model was more muted and pastel-like. The backgrounds look totally different in particular, so we're…
    8875°
  22. #20
    #20
    STABLE DIFFUSION 1.5
    •
    BASE
    VAE Natural Sin Final and last of epiCRealism by epinikion, a realistic base model. Notes from the creator: add "asian, chinese" to negative if you're looking for ethnicities other than Asian Light, shadows, and details are excellent…
    8854°
  23. #21
    #21
    KREA2
    •
    LORA
    by alcaitiff - [KREA 2] Detail Slider
    8851°
  24. #22
    #22
    STABLE DIFFUSION XL 1.0
    •
    BASE
    8849°
  25. #23
    #23
    STABLE DIFFUSION 1.5
    •
    BASE
    Ghibli Diffusion by Nitrosocke was a face-melting anime model of its time, one of the first to cement the idea that Stable Diffusion could one day be stable enough to create films in any distinctive art style, such as that by Japanese art…
    8836°
  26. #24
    #24
    MINIMAX H3
    •
    LORA
    by tsolful - Minimax H3 Turbo Loras larryvrh_4step_pruned
    8835°
  27. #25
    #25
    STABLE DIFFUSION 1.5
    •
    BASE
    8829°
  28. AD
    Keine Installation nötig – KI-Modelle in der Cloud ausführen
    BitVector Prism ist der einfachste Einstieg: Modell wählen, Prompt eingeben und in einer übersichtlichen Web-App generieren, ohne etwas einzurichten. BitVector gibt es auch auf Discord und im Web (SpyGlass).
  29. #26
    #26
    8816°
  30. #27
    #27
    FLUX
    •
    LORA
    Ghibsky Style
    8806°
  31. #28
    #28
    STABLE DIFFUSION 1.5
    •
    LORA
    LoRA designed for removing or adding detail while keeping the style of the render the same. Current strength is good from 0 - 2 depending on your use case. In the future, it can be used as negative weight to reduce details. Works with…
    8805°
  32. #29
    #29
    KREA2
    •
    LORA
    by Loraholic - THE age slider - (Krea-2)
    8789°
  33. #30
    #30
    STABLE DIFFUSION 1.5
    •
    BASE
    Photon by Photographer makes photorealistic images effortlessly. Example prompt from the creator /seed:3883312123 /sampler:dpm2m /size:768x512 /steps:20 /guidance:6 A photo of a young, nerdy woman sitting in a caf, wearing a white shirt…
    8775°
  34. #31
    #31
    STABLE DIFFUSION XL 1.0
    •
    BASE
    von
    tkio
    by tkio - NTR MIX | illustrious-XL | Noob-XL
    8774°
  35. #32
    #32
    STABLE DIFFUSION XL 1.0
    •
    BASE
    8767°
  36. #33
    #33
    KREA2
    •
    LORA
    by HearmemanAI - Realism Enhancer Krea2
    8757°
  37. #34
    #34
    STABLE DIFFUSION 1.5
    •
    EMBEDDING
    Deep Negative Ver. 1.75T by IenML is a popular general quality boosting embedding. Make sure you put it in the negative prompt unless you want your photos to be really disgusting.
    8757°
  38. #35
    #35
    FLUX
    •
    BASE
    Shuttle 3.1 Aesthetic version is a low steps Schnell fine tune by that outperforms the default Schnell model in some cases. Try it with steps 4-6 at guidance 3.5
    8751°
  39. #36
    #36
    8734°
  40. #37
    #37
    STABLE DIFFUSION 1.5
    •
    LORA
    Add More Details - Detail Enhancer / Tweaker (细节调整) LoRA by Lykon writes: his between 0.5 and 1 weight, depending on your preference. You can go lower than 0.5 for a more subtle effect, of course. I also found out that this gives some…
    8730°
  41. #38
    #38
    STABLE DIFFUSION XL 1.0
    •
    BASE
    by GoofiAI - an update on the popular anime model, might be the final release of this. the style is cleaned up, works better with loras
    8729°
  42. #39
    #39
    KREA2
    •
    LORA
    by Loraholic - Skin Detail Slider - (Krea2)
    8725°
  43. #40
    #40
    STABLE DIFFUSION 1.5
    •
    EMBEDDING
    Negative Hand by Nerfgun3 is a hands anatomy fixer that unlike other negative prompts will not alter the overall characteristics of your character. Remember to wrap the trigger inside a negative prompt like this: []
    8724°
  44. #41
    #41
    KREA2
    •
    LORA
    by Loraholic - Wetness Slider - As a always active LoRA: -1.0 to 1.0. Max range before side effects: -2.0 to 2.0
    8713°
  45. #42
    #42
    STABLE DIFFUSION 1.5
    •
    BASE
    AnythingV5/Ink by Yuno779. The harder to use, better for Loras successor to AV3. For most people, you'll want to use AV3 and get nicer pictures. If you're going for accuracy, AV5 might be worth your time. Though iterative but not…
    8709°
  46. #43
    #43
    STABLE DIFFUSION 1.5
    •
    BASE
    Just for fun, this is the original 256x256 original release of Stable Diffusion, released almost two years ago from the CompVis Diffusers repo
    8708°
  47. #44
    #44
    STABLE DIFFUSION 1.5
    •
    BASE
    Dark Sushi Mix brighter version by Aitasai This model is good for traditional anime/cartoon renders without adding any additional dimension to them. Compared to the 25 version, the colors are lighter and brighter. Its style may be compared…
    8708°
  48. #45
    #45
    STABLE DIFFUSION 1.5
    •
    BASE
    Counterfeit2.5 by rqdwdw. Counterfeit 2.5 is the older version of the model, which we liked for its detailed backgrounds and washed watercolor looks. The newer version of Counterfeit is 3.0, which has brighter colors and puts more emphasis…
    8705°
  49. #46
    #46
    ANIMA
    •
    LORA
    By - Motimalu Recommended prompt structure: Positive prompt (quality tags at the start of prompt): masterpiece, best quality, very aesthetic, {{tags}}, {{natural language}}
    8704°
  50. #47
    #47
    STABLE DIFFUSION 1.5
    •
    BASE
    Beautiful Realistic Asians v7 model by pleasebankai
    8702°
  51. #48
    #48
    8697°
  52. #49
    #49
    STABLE DIFFUSION XL 1.0
    •
    BASE
    by GoofyAI - suggested settings is clipskip 1, euler a or dpm2m, guidance 5-10, positive: masterpiece, best quality, absurdres, amazing quality -- negative -- bad quality,worst quality,worst detail,sketch,censored,watermark, signature,…
    8693°
  53. #50
    #50
    STABLE DIFFUSION XL 1.0
    •
    BASE
    8692°
  54. AD
    PirateDiffusion
    Keine Installation nötig – KI-Modelle in der Cloud ausführen
    PirateDiffusion gibt es nur auf Telegram und richtet sich an Profis: tausende Modelle, LoRAs und Workflows per Chat-Befehl, mit unbegrenzter Generierung zum Festpreis.
Heat ist unser interner Trendwert von kühl bis 9999°. Er bewegt sich täglich mit Nutzungs- und Beliebtheitssignalen aus den Communities und Diensten rund um jedes Modell. Die Rangliste umfasst Basismodelle, LoRAs, Embeddings und Textmodelle aller Familien im Katalog.
Nach Modellfamilie stöbern