Ollama
Models Docs Pricing
Sign in Download
Models Download Docs Pricing Sign in
⇅
nsfw · Ollama
Search for models on Ollama.
  • draganis/vanessa

    This is a backup of benevolentjoker/nsfwvanessa:latest

    9,101  Pulls 1  Tag Updated  4 months ago

  • windffooxxyang1130764/mesugaki

    roleplay nsfw

    tools thinking 0.6b 1.7b 4b

    5,206  Pulls 3  Tags Updated  5 months ago

  • windffooxxyang1130764/mesugaki-R18G

    roleplay nsfw zh-cn

    tools thinking 4b

    2,360  Pulls 2  Tags Updated  5 months ago

  • goonsai/qwen2.5-3B-goonsai-nsfw-100k

    To help you with prompts in image and video generation. Same as used on https://goonsai.com Support: r/goonsai/

    31.7K  Pulls 1  Tag Updated  12 months ago

  • jimscard/adult-film-screenwriter-nsfw

    NSFW - Simulates an adult film screenwriter.

    27.3K  Pulls 1  Tag Updated  2 years ago

  • goonsai/qwen2.5-1.5B-goonsai-nsfw-10k

    To help you with prompts in image and video generation. Same as used on https://goonsai.com Support: r/goonsai/

    6,297  Pulls 1  Tag Updated  1 year ago

  • n-inovations/n-ai

    Simple, but powerfull model that almost perfectly simulates a digital friend working more on social behaviour than knowing raw informations, it also loves to learn, he chooses a name of his own

    tools

    31  Pulls 1  Tag Updated  5 days ago

  • jimscard/new-adult-writer

    This is an update of my previous NSFW model (jimscard/adult-film-screenwriter). As the name implies, this model incorporates a system prompt that turns it into an experienced writer of scripts and screenplays for adult films. Use at your own risk! NSFW.

    vision tools thinking audio

    3,116  Pulls 1  Tag Updated  1 month ago

  • AliA456879/small-28m

    Small-28M — a tiny 28.8M parameter pretrained GPT model pretrained from scratch purely on NSFW erotic stories

    1,063  Pulls 1  Tag Updated  1 month ago

  • Nine9/ethicaguide

    “EthicaGuide-7B is a sovereign 7-billion-parameter assistant tuned for local use. It follows the Ethica philosophy: walk beside, not above. No cloud, no tracking, just on-device conversation.”

    167  Pulls 1  Tag Updated  2 months ago

  • MrTails/Tails-assistant-ai-v3.0.0

    search any image, smart, helping with any issues, great for rp(supports NSFW rp), guiders, webX v3.0 support (Network need),

    tools

    6,724  Pulls 1  Tag Updated  9 months ago

  • nishal21/geo-capital-llm

    Based on GPT2 and it sucks and trained on a rented GPU on vast.ai, and still it fucking sucks, well u will get answers, but it may or may not be correct😊

    4  Pulls 1  Tag Updated  2 weeks ago

  • MrTails/Tails-Assistant-ai-v2.5.3

    search any image, smart, helping with any issues, great for rp(supports NSFW rp), suppors WebX v2.5 (Network need),

    tools

    873  Pulls 1  Tag Updated  9 months ago

  • nchapman/l3.1-70b-hanami-x1

    This is an experiment from Sao10k over Euryale v2.2, a model focused on creative roleplay, which worked out nicely.

    70b

    4,624  Pulls 2  Tags Updated  1 year ago

  • smallthinker

    A new small reasoning model fine-tuned from the Qwen 2.5 3B Instruct model.

    3b

    249.5K  Pulls 5  Tags Updated  1 year ago

  • nahiyananwar/mental-health-ai

    Compassionate mental health support AI fine-tuned from Qwen 3 (14B) with 128K context.

    55  Pulls 1  Tag Updated  3 weeks ago

  • Maternion/NuMarkdown-Thinking

    It is the first reasoning OCR VLM. It is specifically trained to convert documents into clean Markdown files, well suited for RAG applications.

    vision thinking 8b

    660  Pulls 2  Tags Updated  5 months ago

  • nick040791/NeuronRouter

    1. Classify the request, -- 2. Reason whether to answer directly, inspect files, run tools, or escalate. -- 3. Choose which worker gets the Task. -- 4. Expose only the minimum tool set.

    vision tools thinking

    147  Pulls 1  Tag Updated  3 months ago

  • rockypod/neotoi-coder

    A Rust / Dioxus 0.7 specialist fine-tuned via RAFT on Qwen3 (3 sizes), optimized for Tailwind v4 + WCAG 2.2 AAA accessibility and dioxus-primitives discipline. Built with CRAFT on a homelab RTX 3090 Ti. No cloud GPUs.

    4b 8b 15b

    79  Pulls 5  Tags Updated  1 month ago

  • nate/instinct

    Instinct is Continue's state-of-the-art open Next Edit model. Robustly fine-tuned from Qwen2.5-Coder-7B, Instinct intelligently predicts your next move to keep you in flow.

    tools

    6,509  Pulls 1  Tag Updated  10 months ago

© 2026 Ollama
Blog Contact