Все модели
Полный список отсортирован по популярности на Replicate.
Exploring the training and use of DreamBooth, with Bart as a subject
Preprompt thy SD spells with ‘zos’ to summon the visionary touch of generative art’s unsung arch-ancestor*, Austin Osman Spare* (1886-1956) *(more info below and in the README!)
Performs speaker identity verification
An SDXL fine-tune based on Pan's Labyrinth
Resize images to specific dimensions with optional aspect ratio preservation.
Depth Any Video with Scalable Synthetic Data
ECCV2022 Quick background removal
A working wsrglow model
Generates in the style of a Funko Pop
Flux lora, use "BLKSCLRA" to trigger image generation
Jennai: an AI Avatar trained via the replicate/fast-flux-trainer
Generates minimal, consistent 3-view design for objects in the style of technical 3D reference sheets. Outputs include front, back, and side views with consistent proportions against a white background.
AI pitch correction that auto-tunes any voice.
Omnilingual ASR 7B by Meta (Unofficial) - Automatic speech recognition supporting 1,693 languages with best-in-class accuracy. Meta's recommended variant for mission-critical transcription requiring maximum quality.
MusicGen Stereo model fine-tuned on the tracks of Oneohtrix Point Never with the text token "OPN"
Fast and cost-efficient SD 1.5 (LCM) model for interior and architectural image generation.
Paintings in the style of selected artists with weights, from the Construction Series of GymDreams8.
ControlNet Scribble for SDXL
Supervised Weakly from hashtAGs
Flux LoRA (owa_face_v1) — a next-gen photorealistic model that fuses cinematic style with the Owa persona, built for immersive storytelling across the Owaverse
super-realistic, TTS speech language model with instant voice cloning
LLAMA-2 7b chat version by Meta. Stream support. Unaltered prompt. Temperature working properly. Economical hardware.
Only makes segmentations for further processing
Microsoft's Florence 2 Base
BGE-M3, the first embedding model which supports multiple retrieval mode, multilingual and multi-granularity retrieval.
staging testing
Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild, using tiled VAE and Juggernaut XL to eliminate plastic skin and smooth textures.
Llama 2 13B fine-tuned on Bob Dylan lyrics
Generate photos of the most handsome man on Earth
An SDXL fine-tune based on Matrix Code art
Classifies text with fine tuned BERT. The model handles batched input texts in a single API call to improve performance and reduce costs.
Video object segmentation for short and long videos
This is a test version, more updates coming
BAAI/bge-reranker-large Model with fp16