Explore

I want to…

Upscale images

Upscaling models that create high-quality images from low-quality images

Restore images

Models that improve or restore images by deblurring, colorization, and removing noise

Flux fine-tunes

Browse the diverse range of fine-tunes the community has custom-trained on Replicate

Make 3D stuff

Models that generate 3D objects, scenes, radiance fields, textures and multi-views.

Latest models

phixtral-2x2_8 is the first Mixure of Experts (MoE) made with two microsoft/phi-2 models, inspired by the mistralai/Mixtral-8x7B-v0.1 architecture

Updated 251 runs

Updated 575 runs

BGE-M3, the first embedding model which supports multiple retrieval mode, multilingual and multi-granularity retrieval.

Updated 234 runs

MetaVoice-1B: 1.2B parameter base model trained on 100K hours of speech

Updated 11.1K runs

Remove background from image

Updated 23.2K runs

incredibly fast whisper using openai/whisper-medium.en NOT the distil model

Updated 356 runs

Qwen1.5 is the beta version of Qwen2, a transformer-based decoder-only language model pretrained on a large amount of data

Updated 3.9K runs

Qwen1.5 is the beta version of Qwen2, a transformer-based decoder-only language model pretrained on a large amount of data

Updated 718 runs

Qwen1.5 is the beta version of Qwen2, a transformer-based decoder-only language model pretrained on a large amount of data

Updated 1.8K runs

Background removal model developed by BRIA.AI, trained on a carefully selected dataset and is available as an open-source model for non-commercial use.

Updated 48.4K runs

Qwen1.5 is the beta version of Qwen2, a transformer-based decoder-only language model pretrained on a large amount of data

Updated 1.4K runs

Qwen1.5 is the beta version of Qwen2, a transformer-based decoder-only language model pretrained on a large amount of data

Updated 719 runs

Qwen1.5 is the beta version of Qwen2, a transformer-based decoder-only language model pretrained on a large amount of data

Updated 58 runs

Updated 174 runs

Remove background from images using BRIA-RMBG-1.4

Updated 10.6K runs

NeverSleep's MiquMaid v1 70B Miqu Finetune, GGUF Q3_K_M quantized by NeverSleep.

Updated 14.1K runs

Base version of Mamba 2.8B, a 2.8 billion parameter state space language model

Updated 237 runs

Base version of Mamba 130M, a 130 million parameter state space language model

Updated 117 runs

Base version of Mamba 370M, a 370 million parameter state space language model

Updated 45 runs

Base version of Mamba 790M, a 790 million parameter state space language model

Updated 44 runs

Base version of Mamba 2.8B Slim Pyjama, a 2.8 billion parameter state space language model

Updated 51 runs

Base version of Mamba 1.4B, a 1.4 billion parameter state space language model

Updated 61 runs

Create a video from an image

Updated 16.3K runs

Merge two images, with an optional third for controlnet.

Updated 5.5K runs

this is a first model

Updated 55 runs

A Visual Language Model for GUI Agents

Updated 2.2K runs

Bokeh Prediction, a hybrid bokeh rendering framework that combines a neural renderer with a classical approach. It generates high-resolution, adjustable bokeh effects from a single image and potentially imperfect disparity maps.

Updated 512 runs

AnimateLCM Cartoon3D Model

Updated 1.3K runs

Finetuned E5 embeddings for instruct based on Mistral.

Updated 131 runs

MoE-LLaVA

Updated 1.4M runs

Updated 127 runs

LLaVA v1.6: Large Language and Vision Assistant (Nous-Hermes-2-34B)

Updated 1.7M runs

LLaVA v1.6: Large Language and Vision Assistant (Vicuna-13B)

Updated 3.3M runs

LLaVA v1.6: Large Language and Vision Assistant (Mistral-7B)

Updated 4.8M runs

one-shot-talking-face-replicate

Updated 1.4K runs

unet clothing segment

Updated 748 runs

Updated 31 runs

Yi-VL-34B is the first open-source 34B VL model worldwide. It demonstrates exceptional performance, ranking first among all existing open-source models in the latest benchmarks including MMMU and CMMMU.

Updated 294 runs

🖼️ Super fast 1.5B Image Captioning/VQA Multimodal LLM (Image-to-Text) 🖋️

Updated 2K runs

High-Quality Image Restoration Following Human Instructions

Updated 10K runs

Updated 1.4K runs

Generates speech from text

Updated 124.8K runs

The Segment Anything Model (SAM) is a powerful and versatile image segmentation model. It leverages a "foundation model" approach, meaning it can be used for various segmentation tasks without needing to be specifically trained for each one.

Updated 283 runs

Source: pipizhao/Pandalyst_13B_V1.0 ✦ Quant: TheBloke/Pandalyst_13B_V1.0-AWQ ✦ Pandalyst: A large language model for mastering data analysis using pandas

Updated 17 runs

A better alternative to SDXL refiners, providing a lot of quality and detail. Can also be used for inpainting or upscaling.

Updated 906.5K runs

'''Last update: Now supports img2img.''' SDXL Canny controlnet with LoRA support.

Updated 733.6K runs

VideoCrafter2: Text-to-Video and Image-to-Video Generation and Editing

Updated 34.2K runs

DiffusionLight: Light Probes by Painting a Chrome Ball

Updated 756 runs

Phi-2 by Microsoft

Updated 2.9K runs

A 70 billion parameter Llama tuned for coding and conversation

Updated 21.8K runs