SEED-Story

Multi-modal Long-form Story Generation Model

EditorRecommendationProductivityArtificial IntelligenceMulti-modal

SEED-Story is a multi-modal long-form story generation model based on a large language model (MLLM). It can generate rich and coherent narrative texts and consistent visual images based on user-provided pictures and text. It represents the cutting-edge technology of artificial intelligence in the fields of creative writing and visual art, with the capability to produce high-quality, multi-modal story content, offering new possibilities for the creative industry.

Generate multi-modal long-form stories by combining text and images to create a coherent narrative.
Based on user input of images and text as the starting point for the story.
Supports the generation of up to 25 multi-modal sequence stories: despite using up to 10 sequences during training.
Maintain image style and character consistency: ensure that generated images match the text's style and character consistency.
StoryStream large-scale dataset: designed for training and benchmarking multi-modal story generation.
Publication of model weights and training code: provides pre-trained Tokenizer
De-Tokenizer
and base model SEED-X.
Supports instruction tuning: further optimizes model performance through instruction adjustment.

The target audience of SEED-Story primarily includes artificial intelligence researchers
data scientists
creative industry professionals
and developers with an interest in artificial intelligence creation. The model aids them in exploring and achieving new forms of creative writing and visual art
and offers a novel approach to content generation.

Generate a multi-modal story based on a specific historical event.
Create personalized children's storybooks by combining user-uploaded images.
Provide initial story concepts and visual designs for movies or games.

1. Clone the SEED-Story repository to the local environment.
2. Install dependencies and set up the Python environment according to requirements.txt.
3. Download and prepare the StoryStream dataset for training and testing.
4. Use the provided script for dataset chunking to enhance training efficiency.
5. Download and save pre-trained model weights to the specified folder.

Visit

SEED-Story Visit Over Time

Monthly Visits

499904316

Bounce Rate

37.31%

Page per Visit

5.8

Visit Duration

00:06:52

SEED-Story Visit Trend

SEED-Story Visit Geography

SEED-Story Traffic Sources

SEED-Story Alternatives

OpenCompass Multi-modal Leaderboard — Real-time updated leaderboard of multi-modal model performance

Productivity•Multi-modal•Performance Evaluation

354

OmniGen — A unified framework for image generation that simplifies multi-task image generation.

Image•Image Generation•Diffusion Models

Mistral-Small-Instruct-2409 — High-performance instruct-tuned AI model

Programming•Artificial Intelligence•Natural Language Processing

xAI — AI to Accelerate Human Scientific Discovery

DeepMind — A leading artificial intelligence research company under Google

Shangchen Zhou — A blog website focused on research and innovation in the fields of computer vision and machine learning.

OpenSource•Computer Vision•Machine Learning

MemoRAG — A memory-based RAG framework for general-purpose applications

Programming•Natural Language Processing•Machine Learning

SuperPrompt — Aimed at helping us understand the engineering of AI agents through prompt design.

Programming•Artificial Intelligence•Machine Learning

RAGLAB — A modular, research-oriented unified framework for retrieval-augmented generation

Programming•Natural Language Processing•Machine Learning

DeepSeek-V2.5 — An AI model integrating general and programming capabilities.

Productivity•Artificial Intelligence•Machine Learning

126

West Lake AI Model — A multimodal model with high emotional and intellectual intelligence

ChineseSelection•Artificial Intelligence•Multimodal

120

RWKV v6 Finch 14B — RWKV v6 Finch 14B is an open-source large model that efficiently processes long texts.

Productivity•Artificial Intelligence•Natural Language Processing

162

C4AI Command R+ 08-2024 — A multilingual advanced dialogue generation model with 104 billion parameters.

Productivity•Artificial Intelligence•Machine Learning

LLM Playground — An interactive platform for loading and testing large language models.

Productivity•Artificial Intelligence•Machine Learning

Boden Annotation Service Enhancement (BASE) — Professional data solution provider

ChineseSelection•Artificial Intelligence•Data Annotation

354

Data-Juicer — A one-stop data processing system that provides high-quality data for large language models.

Productivity•Machine Learning•Data Science

174

labelU-Kit — A front-end annotation component library that supports multiple data annotation methods.

Programming•Data Annotation•Machine Learning

186

Husky-v1 — An open-source language agent that tackles complex multi-step reasoning tasks.

Programming•Artificial Intelligence•Machine Learning

IncarnaMind — Engage in conversations with your documents using large language models like GPT.

Productivity•Machine Learning•PDF

120

Meta Chameleon — An advanced machine learning model that supports non-commercial research.

Education•Machine Learning•Artificial Intelligence

144

VideoTetris — An innovative framework for text-to-video generation

Video•Text-to-Video•Video Generation

768

Video-MME — The first comprehensive benchmark for evaluating the performance of Multi-Modal Large Language Models (MLLMs) in video analysis.

Video•Multi-modal•Video Analysis

438

DIAMOND — A reinforcement learning agent trained in a diffusion world model

Productivity•Machine Learning•Reinforcement Learning

Best AI Websites & Tools

SEED-Story

SEED-Story Visit Over Time

SEED-Story Visit Trend

SEED-Story Visit Geography

SEED-Story Traffic Sources

SEED-Story Alternatives

OpenCompass Multi-modal Leaderboard — Real-time updated leaderboard of multi-modal model performance

OmniGen — A unified framework for image generation that simplifies multi-task image generation.

Mistral-Small-Instruct-2409 — High-performance instruct-tuned AI model

xAI — AI to Accelerate Human Scientific Discovery

DeepMind — A leading artificial intelligence research company under Google

Shangchen Zhou — A blog website focused on research and innovation in the fields of computer vision and machine learning.

MemoRAG — A memory-based RAG framework for general-purpose applications

SuperPrompt — Aimed at helping us understand the engineering of AI agents through prompt design.

RAGLAB — A modular, research-oriented unified framework for retrieval-augmented generation

DeepSeek-V2.5 — An AI model integrating general and programming capabilities.

West Lake AI Model — A multimodal model with high emotional and intellectual intelligence

RWKV v6 Finch 14B — RWKV v6 Finch 14B is an open-source large model that efficiently processes long texts.

C4AI Command R+ 08-2024 — A multilingual advanced dialogue generation model with 104 billion parameters.

LLM Playground — An interactive platform for loading and testing large language models.

Boden Annotation Service Enhancement (BASE) — Professional data solution provider

Data-Juicer — A one-stop data processing system that provides high-quality data for large language models.

labelU-Kit — A front-end annotation component library that supports multiple data annotation methods.

Husky-v1 — An open-source language agent that tackles complex multi-step reasoning tasks.

IncarnaMind — Engage in conversations with your documents using large language models like GPT.

Audio Transcription Tool — Fast, Accurate, and Free Audio to Text Service

UltraEdit — Large-Scale Image Editing Dataset

ControlNeXt — Controllable video and image generation technology

Thousand Brains Project — An AI project exploring brain intelligence

ShareGPT4Video — Enhance AI models for video understanding and generation.

L4GM — 4D Model Reconstruction for Rapid Animation Object Generation

Meta Chameleon — An advanced machine learning model that supports non-commercial research.

VideoTetris — An innovative framework for text-to-video generation

Video-MME — The first comprehensive benchmark for evaluating the performance of Multi-Modal Large Language Models (MLLMs) in video analysis.

DIAMOND — A reinforcement learning agent trained in a diffusion world model