What is the difference between a foundation model and an application built on one?

A foundation model is the underlying trained system — the weights, checkpoint, or API that generates text, images, or other outputs. Applications built on top of it add a user interface, specific prompting, and workflow logic for a particular use case. This category of 268 listings covers the models themselves, such as DeepSeek-R1, GPT-5, and Stable Diffusion XL, rather than the applications built around them.

What are open-weight models and how do they differ from proprietary APIs?

Open-weight models like DistilGPT2 and DeepSeek-R1 release their trained weights publicly, allowing developers to download, fine-tune, and self-host the model without depending on an external API. Proprietary models like those from OpenAI are accessed via an API where you pay per token but never see the underlying weights. Open-weight models offer more control and lower ongoing costs at scale, while proprietary APIs are easier to get started with and receive continuous updates.

How do multimodal vision-language models differ from text-only LLMs?

Multimodal models accept both images and text as input — you can show the model a photo and ask questions about it, have it describe an image, or use visual context alongside written instructions. Text-only LLMs process and generate language only. Many frontier models available through APIs now support multimodal input, though the quality of visual understanding varies across models.

How do I choose the right LLM for a specific task?

The right choice depends on the task type (reasoning, coding, creative writing, vision), latency requirements, context window size, cost per token, and whether you need to fine-tune or self-host. Tools like Roboflow Computer Vision and Firecrawl in this category are examples of specialized models or infrastructure serving specific developer use cases. Evaluating models on a representative sample of your actual task is more reliable than relying on benchmark scores alone.

Best AI Models & LLMs AI Tools

271 tools

Discover the best AI Models & LLMs — from AI Models and AI Apps to AI Hugging Face, AI GitHub Repos, and AI Platforms, this is where developers and researchers discover tools that elevate their projects: training computer vision systems, deploying language models, experimenting with reasoning algorithms, and managing model versions. Browse 271 AI Models & LLMs AI Tools across AI Models, AI Apps, AI Hugging Face, AI GitHub Repos and AI Platforms and more.

271 tools

MPNet is a novel pre-training method for language understanding tasks, improving upon BERT and XLNet. It offers a unified implementation for various pre-training models and…

AI Models & LLMs

StyleSwin is a transformer-based Generative Adversarial Network (GAN) designed for high-resolution image generation. It leverages Swin transformers and a novel double attention…

AI Models & LLMs

Decision Transformer reframes reinforcement learning as a sequence modeling problem, leveraging Transformer architectures like GPT-x and BERT. It generates optimal actions by…

AI Models & LLMs

This GitHub repository provides the official implementation for the NeurIPS 2021 paper 'Revisiting Deep Learning Models for Tabular Data.' It explores deep learning architectures…

AI Models & LLMs

SPP_net is a re-implementation of the Spatial Pyramid Pooling algorithm for deep convolutional networks in visual recognition. It aims to reproduce object detection results from a…

AI Models & LLMs

The ShuffleNet Series repository offers a collection of highly efficient convolutional neural network models designed for mobile devices and resource-constrained environments. It…

AI Models & LLMs

MnasNet automates the design of mobile machine learning models by incorporating speed constraints directly into the search reward function. This platform-aware approach identifies…

AI Models & LLMs

This repository provides an implementation of ConvMixer, a convolutional neural network architecture for image recognition tasks. It is based on the paper "Patches Are All You…

AI Models & LLMs

RepVGG is a powerful and simple ConvNet architecture that achieves high accuracy on ImageNet. It utilizes a VGG-style design with re-parameterization techniques, allowing for…

AI Models & LLMs

MobileOne is an efficient neural network backbone designed for mobile devices. It optimizes for inference speed, achieving under 1 ms on an iPhone12 with high accuracy. This model…

AI Models & LLMs

Vision GNN (ViG) is a PyTorch implementation of Vision Graph Neural Networks developed by Huawei Noah's Ark Lab. It offers efficient AI backbones for image processing tasks,…

AI Models & LLMs

Deformable Convolutional Networks (DCN) enhance deep learning models by adaptively learning sampling offsets for convolutional kernels. This allows networks to better capture…

AI Models & LLMs

PointPillars for KITTI object detection is a GitHub repository providing code to reproduce results from the PointPillars paper on the KITTI dataset. It's a fork of the SECOND…

AI Models & LLMs

SECOND PyTorch is an open-source implementation of the SECOND object detection network for KITTI and NuScenes datasets. It utilizes sparse convolutions for efficient 3D object…

AI Models & LLMs

CTRL is a 1.6 billion-parameter conditional transformer language model for controllable text generation. It conditions on control codes to specify domain, entities, and…

AI Models & LLMs

DialoGPT is a large-scale pretrained language model for dialogue response generation. Developed by Microsoft, it leverages GPT-2 architecture and is trained on extensive Reddit…

AI Models & LLMs

Meena is a 2.6 billion parameter neural conversational model designed for open-domain dialogue. It aims to provide more sensible and specific responses than existing chatbots,…

AI Models & LLMs

GODEL is a large-scale pre-trained Transformer-based model for goal-directed dialog generation. It excels at response generation grounded in external text, enabling efficient…

AI Models & LLMs

RETRO (Retrieval Enhanced Transformers) is an AI model that augments transformer architectures with retrieval capabilities. It accesses a vast database of text passages, including…

AI Models & LLMs

BERT offers two multilingual models: Cased and Uncased, supporting over 100 languages. The Cased model is recommended for non-Latin alphabets and general use, while the Uncased…

AI Models & LLMs

LS-GAN, or Loss-Sensitive Generative Adversarial Networks, is a project focused on advancing GANs. It introduces a novel approach to loss functions, aiming for improved generation…

AI Models & LLMs

This GitHub repository provides an official Chainer implementation for conditional image generation. It utilizes spectral normalization and a projection discriminator for…

AI Models & LLMs

This repository provides a PyTorch implementation of MADE (Masked Autoencoder Density Estimation). It enables turning auto-encoders into autoregressive density models by masking…

AI Models & LLMs

Lyra is a novel, very low-bitrate speech codec developed by Google. It leverages machine learning to compress voice signals, enabling high-quality audio communication even on the…

AI Models & LLMs

SoundStorm is an AI model for efficient, non-autoregressive audio generation. It produces high-quality audio two orders of magnitude faster than previous methods, maintaining…

AI Models & LLMs

GluonTS is a Python package for probabilistic time series modeling, leveraging deep learning models built on PyTorch and MXNet. It facilitates accurate forecasting by generating…

AI Models & LLMs

LSTNet is a deep learning model designed for time series forecasting. It effectively models both long-term and short-term temporal patterns using a combination of recurrent neural…

AI Models & LLMs

Autoformer is an AI model for long-term time series forecasting, leveraging decomposition transformers and auto-correlation mechanisms. It achieves state-of-the-art results across…

AI Models & LLMs

The Informer2020 GitHub repository provides the PyTorch implementation for the Informer model, designed for efficient long sequence time-series forecasting. It features ProbSparse…

AI Models & LLMs

FEDformer is a Frequency Enhanced Decomposed Transformer designed for efficient long-term time series forecasting. It achieves linear complexity with sequence length,…

AI Models & LLMs

PatchTST is an official implementation of a Transformer-based model for long-term time series forecasting. It segments time series into patches and processes them independently,…

AI Models & LLMs

LTSF-Linear is an official PyTorch implementation for "Are Transformers Effective for Time Series Forecasting?". It introduces a family of linear models (Linear, DLinear, NLinear)…

AI Models & LLMs

TimesNet is a foundation model for general time series analysis, transforming 1D series into 2D space for enhanced modeling. It achieves state-of-the-art results in forecasting,…

AI Models & LLMs

This repository provides code for reinforcement learning experiments detailed in the "Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks" paper. It enables…

AI Models & LLMs

This repository contains the code for the NeurIPS 2017 paper "Prototypical Networks for Few-shot Learning." It provides an implementation of a novel approach to few-shot learning,…

AI Models & LLMs

This repository provides PyTorch code for the CVPR 2018 paper 'Learning to Compare: Relation Network for Few-Shot Learning'. It focuses on the Few-Shot Learning aspect, offering…

AI Models & LLMs

This GitHub repository contains the code for the paper "Generative Adversarial Imitation Learning." It implements Trust Region Policy Optimization and provides scripts for…

AI Models & LLMs

This GitHub repository contains the code for the "When to Trust Your Model: Model-Based Policy Optimization" paper. It provides implementations for model-based policy optimization…

AI Models & LLMs

This repository contains experiment code for "Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models." It implements the PETS algorithm, combining…

AI Models & LLMs

PlaNet is a model-based reinforcement learning algorithm that plans from pixels by learning latent dynamics. It efficiently predicts future rewards in a learned latent space,…

AI Models & LLMs

OpenAI Five is a team of five neural networks developed by OpenAI. This AI system has demonstrated the capability to compete against and defeat amateur human teams in the complex…

AI Models & LLMs

AlphaStar is an AI model that achieved Grandmaster level in StarCraft II, a complex real-time strategy game. It utilizes multi-agent reinforcement learning and imitation learning,…

AI Models & LLMs

Policy gradient methods are a class of reinforcement learning algorithms that directly learn a policy function. Unlike value-based methods, they optimize a policy's parameters to…

AI Models & LLMs

SARSA is a reinforcement learning algorithm for learning Markov decision process policies. It updates Q-values based on the agent's current state, action, reward, next state, and…

AI Models & LLMs

Q-learning is a model-free reinforcement learning algorithm that trains an agent to assign values to actions based on current states. It optimizes decision-making by maximizing…

AI Models & LLMs

This project provides a Keras-based implementation of Relational Graph Convolutional Networks (R-GCNs). It is designed for semi-supervised node classification on directed…

AI Models & LLMs

CompGCN is a PyTorch implementation of Composition-Based Multi-Relational Graph Convolutional Networks. It enables advanced graph representation learning for tasks like knowledge…

AI Models & LLMs

This repository provides a PyTorch implementation of the "Simple and Deep Graph Convolutional Networks" (GCNII) model. It includes code for replicating semi-supervised and…

AI Models & LLMs

This GitHub repository provides TensorFlow and PyTorch implementations for the PPNP and APPNP models, introduced in the "Predict then Propagate: Graph Neural Networks meet…

AI Models & LLMs

This repository provides the official implementation for the "Simplifying Graph Convolutional Networks" paper. It offers a streamlined Graph Convolutional Network (GCN) model that…

AI Models & LLMs

Graph Neural PDEs is an open-source project providing Python code for Graph Neural Networks (GNNs) that treat graph learning as a continuous diffusion process. It implements GRAND…

AI Models & LLMs

This blog post introduces a new PyTorch Hopfield layer based on modern Hopfield Networks. It generalizes classical Hopfield Networks to continuous states and patterns, offering…

AI Models & LLMs

A self-organizing map (SOM) is an unsupervised machine learning technique that reduces data dimensionality while preserving topological structure. It creates a low-dimensional…

AI Models & LLMs

A radial basis function network is an artificial neural network utilizing radial basis functions as activation functions. It's adept at function approximation, time series…

AI Models & LLMs

A Bayesian network is a probabilistic graphical model representing variables and their conditional dependencies using a directed acyclic graph. It's ideal for predicting the…

AI Models & LLMs

Markov Random Fields (MRFs) are undirected graphical models representing probabilistic relationships between random variables. They are crucial in AI for modeling dependencies in…

AI Models & LLMs

SAGPool is an official PyTorch implementation of Self-Attention Graph Pooling, presented at ICML 2019. This read-only, archived repository provides the code for a novel graph…

AI Models & LLMs

Temporal Graph Networks (TGNs) offer a generic, efficient framework for deep learning on dynamic graphs. This Python-based library utilizes novel memory modules and graph-based…

AI Models & LLMs

EvolveGCN provides code for Evolving Graph Convolutional Networks designed for dynamic graphs. This AI model, published in AAAI 2020, offers a framework for analyzing and modeling…

AI Models & LLMs

xDeepFM is an open-source deep learning toolkit for factorization models, specifically releasing the source code for the eXtreme Deep Factorization Machine (xDeepFM) model. It…

AI Models & LLMs

This GitHub repository provides a TensorFlow implementation of the Attentional Factorization Machine (AFM) model. It allows users to train and evaluate AFM for tasks like…

AI Models & LLMs

LightGCN is a simplified Graph Convolutional Network designed for recommendation systems. It focuses on the essential neighborhood aggregation component for collaborative…

AI Models & LLMs

Neural Graph Collaborative Filtering (NGCF) is a recommendation framework leveraging graph neural networks. It explicitly encodes collaborative signals through high-order…

AI Models & LLMs

This AI model introduces Poincaré embeddings for learning hierarchical representations of symbolic data. It embeds data into hyperbolic space, capturing both hierarchy and…

AI Models & LLMs

Cohere's comprehensive API documentation empowers developers to seamlessly integrate advanced natural language processing and generation capabilities into their applications.…

AI Models & LLMs

BioGPT is a generative pre-trained transformer model for biomedical text generation and mining. Developed by Microsoft, it offers pre-trained and fine-tuned checkpoints for…

AI Models & LLMsHealthcare & Life Sciences

StarCoder is a large language model trained on source code and natural language. It excels at code generation, completion, and text generation tasks. This repository provides…

AI Models & LLMs

LLaVA is a visual instruction tuning model that combines large language and vision capabilities. It aims to achieve GPT-4V level performance, enabling multimodal understanding and…

AI Models & LLMs

Kandinsky 2 is a multilingual text-to-image latent diffusion model. It offers advanced image generation capabilities, including text-to-image, image-to-image, and inpainting. The…

AI Models & LLMs

Qwen3 is a series of large language models developed by the Qwen team at Alibaba Cloud. It offers enhanced capabilities in instruction following, reasoning, text comprehension,…

AI Models & LLMs

The Anthropic SDK for Python provides developers with direct access to the Claude API. This open-source library facilitates integration of advanced AI capabilities into Python…

AI Models & LLMs

StableLM is an open-source language model series developed by Stability AI. This GitHub repository hosts ongoing development, providing access to various checkpoints like…

AI Models & LLMs

Open-sourced code for MiniGPT-4 and MiniGPT-v2, advanced large language models enhancing vision-language understanding. These models enable multi-task learning for vision-language…

AI Models & LLMs

AUI's Apollo-1 is a neuro-symbolic foundation model for task-oriented dialogue. It reasons in code, not weights, enabling companies to build agents that follow business rules…

AI Models & LLMs

Thunder Compute offers the world's cheapest GPUs, providing access to NVIDIA H100s at competitive hourly rates. Their cloud platform utilizes advanced GPU virtualization to make…

AI Models & LLMs

NinjaTools offers an all-in-one AI workspace with dozens of professional tools for a single subscription. Access AI search, content generation, image and video creation, and PDF…

AI Models & LLMs

Text Generator offers a unified API for advanced AI text, vision, and speech capabilities. It focuses on strong output quality, privacy-first infrastructure, and predictable…

AI Models & LLMs

Access over 600 AI models through a single API, including GPT-5.5, Claude Opus 4.7, and Gemini 3.5. This platform offers a cost-effective solution, potentially saving up to 80%…

AI Models & LLMs

MimicPC is an open-source, customizable, and affordable AI platform for creating images, videos, and audio. Train LoRA models, add custom models, and leverage popular AI tools…

AI Models & LLMs

Kie.ai offers a unified API to access over 100 leading AI models for video, image, audio, and LLM generation. It provides stable performance, transparent pricing, and significant…

AI Models & LLMs

Arcee AI is a US-based open intelligence lab developing advanced, open-weight AI models. Their Trinity family of models offers state-of-the-art performance for enterprise…

AI Models & LLMs

Upstage builds enterprise-grade large language models and document-processing engines, including Document Parse, Information Extract, and the Solar LLM family, to automate…

AI Models & LLMsInsurance

A crowdsourced benchmark where users create with top AI models and vote to rank them by design quality. Built for anyone comparing AI models for websites, apps, games, images, and…

AI Models & LLMs

deAPI is a unified API for open-source generative AI models, covering image generation, text-to-speech, transcription, and more. Requests run on a decentralized GPU cloud,…

AI Models & LLMs

An omni-modal AI gateway and API platform with in-house foundation models for text, image, and video, plus consumer apps for chatting, creating, and building agentic tools.

AI Models & LLMs

A free tool that detects your GPU or Mac and shows which open AI models, including LLMs, coding, image, and video models, you can run locally, with VRAM requirements and speed…

AI Models & LLMs

APIMart is a unified AI API platform that provides access to over 500 AI models, including GPT-5 and Claude. It offers transparent pricing with no hidden fees, making it easy for…

AI Models & LLMs

A unified generative AI API platform that lets developers run leading video, image, audio, and LLM models such as FLUX, Kling, Seedance, and Nano Banana through one consistent,…

AI Models & LLMs

A model access layer that gives AI teams one API for 170+ LLM, image, video, and audio models, with documented capabilities, transparent pricing, smart routing, and 99.9% uptime…

AI Models & LLMs

A unified AI gateway that gives developers 100+ models such as GPT, Claude, Gemini, DeepSeek, Kimi, Qwen, and GLM through one API key, with OpenAI/Anthropic/Gemini compatibility…

AI Models & LLMs

Kie.ai is a unified AI API that gives developers one endpoint for leading video, image, music, and LLM models at lower cost than official APIs. It is built for teams shipping…

AI Models & LLMs

legml.ai builds sovereign, sector-specialized large language models and a private voice AI meeting agent for French and European enterprises, deployed on-premise with…

AI Models & LLMsLegal Services

A full-stack sovereign AI platform that lets companies and governments run private LLMs on their own GPU infrastructure, with a terminal-native AI coding assistant and memory…

AI Models & LLMsGovernment & Public Sector

ZenMux is an enterprise LLM platform offering a unified API to 100+ AI models such as Claude, GPT, and Gemini, with intelligent routing, developer observability, and AI Model…

AI Models & LLMs

An open-source, OpenAI-compatible API gateway that routes requests across 200+ models from 40+ providers through one endpoint, with real-time cost and latency analytics, caching,…

AI Models & LLMs

PoYo.AI is a unified AI API platform that gives developers one API key to access 500+ image, video, music, chat, and 3D models with credit-based, pay-as-you-go pricing.

AI Models & LLMs

An open-source model community and platform for exploring, running, fine-tuning, and deploying AI models and datasets, with a Python library and hosted studios for building AI…

AI Models & LLMs

A high-performance AI API aggregation platform that provides unified access to 100+ top AI models for chat, image, video, and music through one OpenAI-compatible API, with…

AI Models & LLMs

A free browser demo of OpenAI's open-weight models, gpt-oss-120b and gpt-oss-20b, that lets developers try the models and adjust the reasoning level.

AI Models & LLMs

A free online demo of the open-source DeepSeek V3 model (671B parameters) that runs in your browser with no registration, plus links to run the model locally for personal and…

AI Models & LLMs

Claude API provides access to the complete Claude model family through a single, reliable API proxy. With a one-minute setup, 99.9% uptime, and 24/7 developer support, it ensures…

AI Models & LLMs

APIMODELS is a multi-model AI API platform that aggregates over 110 models, including Claude and GPT, offering them at prices up to 95% lower than official APIs. It provides a…

AI Models & LLMs

you.bot provides a unified API for accessing over 80 AI models, offering significant savings of up to 90% compared to direct provider rates. Users can pay only for successful…

AI Models & LLMs

Catnip.AI is an AI research company specializing in real-time multimodal intelligence, enabling models to see, listen, speak, and respond in real time, enhancing interactive…

AI Models & LLMsEducation & E-learning

discode.ai offers a unique AI experience where users choose the rhythm, not the algorithm. With over 100 AI models, it provides ecological, private, and smart solutions tailored…

AI Models & LLMs

Poolside is a foundation model company focused on creating artificial general intelligence. They develop open-weight foundation models optimized for quality, speed, and…

AI Models & LLMs

MiniMax is a leading global AI technology company focused on co-creating intelligence with everyone. It develops a series of multimodal AI models and offers a range of AI-native…

AI Models & LLMs

OfoxAI offers access to over 100 AI models, including GPT-5.5, Claude Opus 4.8, and Gemini 3.5 Flash, through a single API. It provides real-time pricing and is compatible with…

AI Models & LLMs

Sup AI was a platform focused on orchestrating an ensemble of AI models for improved accuracy. Although the company is winding down, the mission continues with a focus on causal…

AI Models & LLMs

Nous Research is a leader in the American open source AI movement, specializing in training open source language models and building infrastructure for unbiased training. Their…

AI Models & LLMs

Falcon LLM is a generative large language model designed to advance applications and use cases, making advanced AI accessible and impactful across various industries and…

AI Models & LLMs

Claude Fable 5 is a cutting-edge AI model designed for complex, long-running tasks in coding and knowledge work. It offers advanced capabilities with robust safeguards for…

AI Models & LLMsFinance & Banking

Mira Murati is a prominent AI influencer, currently leading at Thinking Machines. Previously, she served as CTO at OpenAI. Her work focuses on open-weight AI models and multimodal…

AI Models & LLMs

Alexandr Wang is the Chief AI Officer at Meta and founder of Meta Superintelligence Labs and Scale AI. He is known for his work on AI models like Muse Spark 1.2 and Muse Glimmer,…

AI Models & LLMs

Aidan Gomez is the cofounder of Cohere, a company focused on advancing AI technology. He shares insights and updates on AI developments and industry trends on his X profile.

AI Models & LLMs

Arthur Mensch is the Co-founder and CEO of Mistral AI. He shares insights on AI development, open-source models, and strategic partnerships, focusing on ensuring global benefits…

AI Models & LLMs

OpenRouter Discord is a community platform where users can discover and use the latest LLMs. With over 500 models, including 50+ free options, it offers explorable data,…

AI Models & LLMs

Open Source LLMs Discord is a community for discussing and supporting AI Large Language Models. With over 18,000 members, it provides a platform for enthusiasts and experts to…

AI Models & LLMs

Mistral AI is a Paris-based company developing advanced large language models. This GitHub repository provides an independent third-party profile of its public API surface,…

AI Models & LLMs

Burrito Core is an inference harness for gpt-oss, offering compatibility with OpenAI and Anthropic APIs. It supports native Python and browser tools, utilizing llama.cpp or vLLM…

AI Models & LLMs

Korean LLM v3 is a GitHub project that enables building and running a Korean-specialized 1.09B-parameter language model from scratch. It utilizes PyTorch for development and…

AI Models & LLMs

Frequently Asked Questions

A foundation model is the underlying trained system — the weights, checkpoint, or API that generates text, images, or other outputs. Applications built on top of it add a user interface, specific prompting, and workflow logic for a particular use case. This category of 268 listings covers the models themselves, such as DeepSeek-R1, GPT-5, and Stable Diffusion XL, rather than the applications built around them.
Open-weight models like DistilGPT2 and DeepSeek-R1 release their trained weights publicly, allowing developers to download, fine-tune, and self-host the model without depending on an external API. Proprietary models like those from OpenAI are accessed via an API where you pay per token but never see the underlying weights. Open-weight models offer more control and lower ongoing costs at scale, while proprietary APIs are easier to get started with and receive continuous updates.
The right choice depends on the task type (reasoning, coding, creative writing, vision), latency requirements, context window size, cost per token, and whether you need to fine-tune or self-host. Tools like Roboflow Computer Vision and Firecrawl in this category are examples of specialized models or infrastructure serving specific developer use cases. Evaluating models on a representative sample of your actual task is more reliable than relying on benchmark scores alone.

Know a standout that belongs in AI Models & LLMs?

ToolPotion is the AI tools directory people browse to find the best AI Models & LLMs. Listing is free and every submission is editor-reviewed — put your tool where the search starts.

Submit an AI Tool for AI Models & LLMs