SuperAPI AI Directory
Partner with us中文AI newsFavorites

All AI tools

35,738 results · Page 722 of 993

Korewa AI

AI assistants

Korewa AI is an AI chat platform aimed at anime fans, focused on visual novel-style character animation and lifelike conversations with anime characters. Users can chat with user-generated anime characters on the platform, or create their own characters: upload character images, write memories, personality traits, and expressions, allowing characters to display emotions visually. Officially, its AI text model is said to be fine-tuned for anime characters, able to simulate emotions, let characters remember conversation content, and gradually evolve through interaction. The platform emphasizes its anime vertical positioning and visual experience, claiming to provide powerful and emotionally rich character simulation models on the market. After creating a character, users can chat with it and observe the effect of it "coming to life." Specific fees, pricing, platform support, and usage restrictions are not currently provided.

Read more

Krea 2

AI images

Krea 2 is the first image foundation model built from scratch by Krea AI, released in 2026. The official technical report describes it as a 12.9-billion-parameter single-stream diffusion Transformer that uses Qwen3-VL as its text encoder, emphasizes aesthetic diversity and creative control, and combines text-to-image generation with prompt expansion and visual style references. This website, krea-2.pro, is an independent workspace, not an official Krea AI site. Around Krea 2 it offers processes such as exploratory text-to-image generation, style references, mood boards, creativity adjustment, and rapid visual direction iteration, and features annual credit packages.

Read more

Krea 2 Turbo

AI images

Krea 2 Turbo is a distilled, accelerated version of the Krea 2 base model. It uses eight-step sampling to generate a 1024px draft in about two seconds, after which a 2K image can be output. It supports text-to-image, image-to-image, and LoRA style control. Community or self-trained LoRAs can be attached as a bidirectional slider to adjust style strength without increasing the cost of a single render. The model focuses on photographic performance, with natural depth of field, realistic materials, and light behavior. The base model is a 12-billion-parameter diffusion Transformer, with sampling compressed to eight steps through trajectory distillation. Its weights are released under a custom license that allows commercial use. This website wraps it as a browser-based AI image generator, with no need to configure a local GPU. It offers free credits, and paid plans provide 500, 2,000, or 6,000 credits per month, or 6,000, 24,000, and 72,000 credits accumulated respectively on annual billing. It supports PNG, JPG, and WebP export and priority queue rendering.

Read more

Krea 3

AI images

Krea 3 is a standalone AI image editor focused on image-to-image editing of existing photos: after uploading a photo, enter a text description to change the lighting, color, background, or a single object while keeping faces and product appearances as recognizable as possible. It supports established models such as GPT Image 2, Nano Banana Pro, and Seedream, and offers features including multi-reference image merging, video generation, over 600 style presets, a My Creations library, and 1K/2K/4K export. This website is not affiliated with Krea AI officially and does not provide Krea's own model weights.

Read more

KRЯRL DЯAWINGS: Runway ML -- 3rd "Model" (based on long poses)

AI images

This is a personal blog post that records the author's experimental process and reflections on using Runway ML to train their third drawing generation model (based on long-pose sketches). The author trained for over 3,000 steps using 348 of their own sketches, ultimately achieving an FID score of 80.13, and discussed the impact of training steps and dataset expansion (such as adding works by Egon Schiele) on the style and realism of the output results. The article is a record of a creative experiment, not official Runway ML documentation or a product introduction.

Read more

kuakua

AI learning

Kuakua is an online learning and self-assessment platform for mental health centered on positive psychology and mindfulness practice, offering AI-driven psychological tools, mindfulness exercises, and a happiness generator. The website brings together a rich library of psychology resources, covering topics such as depression tests, personality tests, fun psychological tests, clinical psychological tests, autism spectrum tests, neurodiversity tests, and family and child development tests; it also provides practical assessment tools such as child development milestone tracking, parenting risk self-checks, and kindergarten school readiness checks, combined with official CDC checklists and scale guidelines. The platform also includes memory, math, and logical thinking games, visualized experiences of classic psychology experiments, and interactive quizzes on topics such as working memory and cognitive control. Users can install a Chrome or Edge browser extension to receive daily mindfulness reflection questions while browsing the web. The website emphasizes that happiness exists in the present moment, achieving physical and mental balance through the combination of knowledge and action.

Read more