Open Source Python AI Models - Page 7
Sort By:
Python AI Models
View 806 business solutionsBrowse free open source Python AI Models and projects below. Use the toggles on the left to filter open source Python AI Models by OS, license, language, programming language, and project status.
-
MongoDB Atlas runs apps anywhereMongoDB Atlas gives you the freedom to build and run modern applications anywhere—across AWS, Azure, and Google Cloud. With global availability in over 115 regions, Atlas lets you deploy close to your users, meet compliance needs, and scale with confidence across any geography.
-
Build Data Resilience - Take the Assessment TodayIs your recovery strategy as strong as you think? Take this quick self-assessment to check your recovery readiness and gain tailored insights. In only 2 minutes, you'll learn where you fall on the recovery readiness scale.
-
1
HunyuanWorld-Voyager
RGBD video generation model conditioned on camera input
HunyuanWorld-Voyager is a next-generation video diffusion framework developed by Tencent-Hunyuan for generating world-consistent 3D scene videos from a single input image. By leveraging user-defined camera paths, it enables immersive scene exploration and supports controllable video synthesis with high realism. The system jointly produces aligned RGB and depth video sequences, making it directly applicable to 3D reconstruction tasks. At its core, Voyager integrates a world-consistent video diffusion model with an efficient long-range world exploration engine powered by auto-regressive inference. To support training, the team built a scalable data engine that automatically curates large video datasets with camera pose estimation and metric depth prediction. As a result, Voyager delivers state-of-the-art performance on world exploration benchmarks while maintaining photometric, style, and 3D consistency. -
2
LaMDA-pytorch
Open-source pre-training implementation of Google's LaMDA in PyTorch
Open-source pre-training implementation of Google's LaMDA research paper in PyTorch. The totally not sentient AI. This repository will cover the 2B parameter implementation of the pre-training architecture as that is likely what most can afford to train. You can review Google's latest blog post from 2022 which details LaMDA here. You can also view their previous blog post from 2021 on the model. -
3
Lyra 2
Project Lyra: Open Generative 3D World Models
The Lyra 2 project is a research-driven framework developed by NVIDIA that focuses on building open generative 3D world models using advanced diffusion-based techniques. It enables the creation of fully explorable 3D environments from minimal inputs such as a single image or video, leveraging self-distillation methods to generate consistent spatial representations. The system evolves across versions, with newer iterations introducing long-horizon generation and improved 3D consistency across frames. It combines elements of computer vision, generative modeling, and spatial intelligence to produce dynamic and navigable virtual worlds. The architecture is designed to handle both 3D and 4D scene generation, making it suitable for applications such as simulation, gaming, and virtual environments. By emphasizing open implementations, the project provides researchers and developers with access to cutting-edge generative modeling techniques. -
4
Mask2Former
Code release for "Masked-attention Mask Transformer
Mask2Former is a unified segmentation architecture that handles semantic, instance, and panoptic segmentation with one model and one training recipe. Its core idea is to cast segmentation as mask classification: a transformer decoder predicts a set of mask queries, each with an associated class score, eliminating the need for task-specific heads. A pixel decoder fuses multi-scale features and feeds masked attention in the transformer so each query focuses computation on its current spatial support. This leads to accurate masks with sharp boundaries and strong small-object performance while remaining efficient on high-resolution inputs. The project provides extensive configurations and pretrained models across popular benchmarks like COCO, ADE20K, and Cityscapes. Built on top of Detectron2, it includes training scripts, inference tools, and visualization utilities that make experimentation straightforward. -
AI-generated apps that pass security reviewRetool lets you generate dashboards, admin panels, and workflows directly on your data. Type something like “Build me a revenue dashboard on my Stripe data” and get a working app with security, permissions, and compliance built in from day one. Whether on our cloud or self-hosted, create the internal software your team needs without compromising enterprise standards or control.
-
5
MiMo-V2.5-ASR
Robust Speech Recognition Across Languages, Dialects
MiMo-V2.5-ASR is an advanced automatic speech recognition system developed as part of Xiaomi’s MiMo AI ecosystem. It is designed to handle complex acoustic environments, including noisy conditions and diverse speaker variations. The model supports multiple languages and dialects, enabling robust transcription across global use cases. It leverages modern deep learning architectures to improve accuracy and adaptability in real-world scenarios. The system is built to integrate with broader AI pipelines, including voice assistants and multimodal systems. It focuses on scalability and performance, making it suitable for both research and production applications. Overall, it represents a high-performance speech recognition solution optimized for versatility and reliability. -
6
MiniCPM4
Ultra-Efficient LLMs on End Device
MiniCPM4 is part of the MiniCPM family of ultra-efficient large language models designed specifically for high performance on edge devices and resource-constrained environments. Unlike traditional large-scale models that require extensive computational resources, MiniCPM4 focuses on delivering competitive reasoning and language capabilities while maintaining significantly lower latency and higher efficiency. It achieves this through optimized architectures, scalable training strategies, and techniques such as long-context pretraining and YaRN-based length extension, allowing it to handle sequences up to 128K tokens effectively. The model demonstrates strong performance across tasks such as long-text comprehension, reasoning, and general language generation, often outperforming similar-sized models in both speed and accuracy. MiniCPM4 is available in multiple parameter sizes, making it adaptable to different deployment scenarios ranging from mobile devices to GPUs. -
7
Miso TTS
Miso TTS is an 8 billion, highly emotive text-to-speech model
Miso TTS is an advanced 8-billion-parameter text-to-speech model developed by Miso Labs for generating highly expressive and natural-sounding conversational speech. Built on an RVQ Transformer architecture inspired by Sesame CSM, it combines a powerful Llama-based backbone with an autoregressive audio decoder to produce high-quality audio from text. The model supports both standard speech synthesis and voice-conditioned generation using optional audio prompts for voice cloning. Miso TTS generates Mimi audio codes and can leverage conversation history to create more contextually aware and realistic dialogue. Designed for local deployment, it offers watermarking by default to help promote responsible use of generated audio. With its focus on emotive speech generation, Miso TTS delivers state-of-the-art performance for AI voice applications, virtual assistants, and conversational AI experiences. -
8
MuJoCo MPC
Real-time behaviour synthesis with MuJoCo, using Predictive Control
MuJoCo MPC (MJPC) is an advanced interactive framework for real-time model predictive control (MPC) built on top of the MuJoCo physics engine, developed by Google DeepMind. It allows researchers and roboticists to design, visualize, and execute complex control tasks for simulated or real robotic systems. MJPC integrates a high-performance GUI and multiple predictive control algorithms, including iLQG, gradient descent, and Predictive Sampling — a competitive, derivative-free method that achieves robust real-time control. The system supports multi-shooting optimization, enabling precise motion planning across diverse domains like quadruped locomotion, humanoid tracking, and dexterous manipulation. In addition to its C++ core, MJPC includes an experimental Python API, enabling integration with custom models and MuJoCo tasks for flexible scripting and experimentation. -
9
Multi-Agent Emergence Environments
Environment generation code for the paper "Emergent Tool Use"
multi-agent-emergence-environments is an open source research environment framework developed by OpenAI for the study of emergent behaviors in multi-agent systems. It was designed for the experiments described in the paper and blog post “Emergent Tool Use from Multi-Agent Autocurricula”, which investigated how complex cooperative and competitive behaviors can evolve through self-play. The repository provides environment generation code that builds on the mujoco-worldgen package, enabling dynamic creation of simulated physical environments. Developers can construct custom environments by combining modular components such as Boxes, Ramps, and RandomWalls using a flexible layering approach that reduces code duplication. The framework includes several predefined environments—such as Hide and Seek, Box Locking, Blueprint Construction, and Shelter Construction—that model distinct problem-solving and collaboration scenarios. -
Earn up to 16% annual interest with Nexo.Generate interest, access liquidity without selling, and execute trades seamlessly. All in one platform. Geographic restrictions, eligibility, and terms apply.
-
10
OpenAI Harmony
Renderer for the harmony response format to be used with gpt-oss
Harmony is a response format developed by OpenAI for use with the gpt-oss model series. It defines a structured way for language models to produce outputs, including regular text, reasoning traces, tool calls, and structured data. By mimicking the OpenAI Responses API, Harmony provides developers with a familiar interface while enabling more advanced capabilities such as multiple output channels, instruction hierarchies, and tool namespaces. The format is essential for ensuring gpt-oss models operate correctly, as they are trained to rely on this structure for generating and organizing their responses. For users accessing gpt-oss through third-party providers like HuggingFace, Ollama, or vLLM, Harmony formatting is handled automatically, but developers building custom inference setups must implement it directly. With its flexible design, Harmony serves as the foundation for creating more interpretable, controlled, and extensible interactions with open-weight language models. -
11
PRM800K
800,000 step-level correctness labels on LLM solutions to MATH problem
PRM800K is a process supervision dataset accompanying the paper Let’s Verify Step by Step, providing 800,000 step-level correctness labels on model-generated solutions to problems from the MATH dataset. The repository releases the raw labels and the labeler instructions used in two project phases, enabling researchers to study how human raters graded intermediate reasoning. Data are stored as newline-delimited JSONL files tracked with Git LFS, where each line is a full solution sample that can contain many step-level labels and rich metadata such as labeler UUIDs, timestamps, generation identifiers, and quality-control flags. Each labeled step can include multiple candidate completions with ratings of -1, 0, or +1, optional human-written corrections (phase 1), and a chosen completion index, along with a final finish reason such as found_error, solution, bad_problem, or give_up. -
12
Poetiq
Reproduction of Poetiq's record-breaking submission to the ARC-AGI-1
poetiq-arc-agi-solver is the open-source codebase from Poetiq that replicates their record-breaking submission to the challenging benchmark suite ARC-AGI (both ARC-AGI-1 and ARC-AGI-2). The project demonstrates a system that orchestrates large language models (LLMs) — like those from major providers — with carefully engineered prompting, reasoning workflows, and dynamic strategies, to tackle the abstract, logic-heavy problems in ARC-AGI. Instead of relying on a single prompt or fixed strategy, their solver dynamically adapts the reasoning path, selecting what to ask or analyze next depending on intermediate results — effectively compositing reasoning, perception, and program synthesis (or symbolic manipulation) in a loop. The repository allows others to reproduce their results, experiment with different LLM backends (e.g. the user may supply keys for supported models), and observe how their adaptive meta-system handles the logic and abstraction challenges. -
13
Qwen2-Audio
Repo of Qwen2-Audio chat & pretrained large audio language model
Qwen2-Audio is a large audio-language model by Alibaba Cloud, part of the Qwen series. It is trained to accept various audio signal inputs (including speech, sounds, etc.) and perform both voice chat and audio analysis, producing textual responses. It supports two major modes: Voice Chat (interactive voice only input) and Audio Analysis (audio + text instructions), with both base and instruction-tuned models. It is evaluated on many benchmarks (speech recognition, translation, sound classification, emotion, etc.), and offers pretrained models (e.g. 7B) released via ModelScope and Hugging Face. Code & examples provided with Hugging Face transformers, and usage via AutoProcessor, model classes etc. High performance on many standard benchmarks: ASR, speech-emotion recognition, vocal sound classification, speech translation etc. -
14
Qwen2.5-Omni
Capable of understanding text, audio, vision, video
Qwen2.5-Omni is an end-to-end multimodal flagship model in the Qwen series by Alibaba Cloud, designed to process multiple modalities (text, images, audio, video) and generate responses both as text and natural speech in streaming real-time. It supports “Thinker-Talker” architecture, and introduces innovations for aligning modalities over time (for example synchronizing video/audio), robust speech generation, and low-VRAM/quantized versions to make usage more accessible. It holds state-of-the-art performance in many multimodal benchmarks, particularly spoken language understanding, audio reasoning, image/video understanding, etc. Very strong benchmark performance across modalities (audio understanding, speech recognition, image/video reasoning) and often outperforming or matching single-modality models at a similar scale. Real-time streaming responses, including natural speech synthesis (text-to-speech) and chunked inputs for low latency interaction. -
15
Qwen3-Omni
Qwen3-omni is a natively end-to-end, omni-modal LLM
Qwen3-Omni is a natively end-to-end multilingual omni-modal foundation model that processes text, images, audio, and video and delivers real-time streaming responses in text and natural speech. It uses a Thinker-Talker architecture with a Mixture-of-Experts (MoE) design, early text-first pretraining, and mixed multimodal training to support strong performance across all modalities without sacrificing text or image quality. The model supports 119 text languages, 19 speech input languages, and 10 speech output languages. It achieves state-of-the-art results: across 36 audio and audio-visual benchmarks, it hits open-source SOTA on 32 and overall SOTA on 22, outperforming or matching strong closed-source models such as Gemini-2.5 Pro and GPT-4o. To reduce latency, especially in audio/video streaming, Talker predicts discrete speech codecs via a multi-codebook scheme and replaces heavier diffusion approaches. -
16
Stable Diffusion Version 2
High-Resolution Image Synthesis with Latent Diffusion Models
Stable Diffusion (the stablediffusion repo by Stability-AI) is an open-source implementation and reference codebase for high-resolution latent diffusion image models that power many text-to-image systems. The repository provides code for training and running Stable Diffusion-style models, instructions for installing dependencies (with notes about performance libraries like xformers), and guidance on hardware/driver requirements for efficient GPU inference and training. It’s organized as a practical, developer-focused toolkit: model code, scripts for inference, and examples for using memory-efficient attention and related optimizations are included so researchers and engineers can run or adapt the model for their own projects. The project sits within a larger ecosystem of Stability AI repositories (including inference-only reference implementations like SD3.5 and web UI projects) and the README points users toward compatible components, recommended CUDA/PyTorch versions. -
17
Stable Virtual Camera
Stable Virtual Camera: Generative View Synthesis with Diffusion Models
Stable Virtual Camera is a multi-view diffusion model developed by Stability AI that transforms 2D images into immersive 3D videos with realistic depth and perspective. Unlike traditional methods that require complex reconstruction or scene-specific optimization, this model allows users to generate novel views from any number of input images and define custom camera trajectories, enabling dynamic exploration of scenes. It supports various aspect ratios and can produce 3D-consistent videos up to 1,000 frames, making it a versatile tool for creators seeking to enhance visual storytelling. -
18
Stable-Dreamfusion
Text-to-3D & Image-to-3D & Mesh Exportation with NeRF + Diffusion
A pytorch implementation of the text-to-3D model Dreamfusion, powered by the Stable Diffusion text-to-2D model. This project is a work-in-progress and contains lots of differences from the paper. The current generation quality cannot match the results from the original paper, and many prompts still fail badly! Since the Imagen model is not publicly available, we use Stable Diffusion to replace it (implementation from diffusers). Different from Imagen, Stable-Diffusion is a latent diffusion model, which diffuses in a latent space instead of the original image space. Therefore, we need the loss to propagate back from the VAE's encoder part too, which introduces extra time costs in training. We use the multi-resolution grid encoder to implement the NeRF backbone (implementation from torch-ngp), which enables much faster rendering. -
19
Step-Audio-EditX
LLM-based Reinforcement Learning audio edit model
Step-Audio-EditX is an open-source, 3 billion-parameter audio model from StepFun AI designed to make expressive and precise editing of speech and audio as easy as text editing. Rather than treating audio editing as low-level waveform manipulation, this model converts speech into a sequence of discrete “audio tokens” (via a dual-codebook tokenizer) — combining a linguistic token stream and a semantic (prosody/emotion/style) token stream — thereby abstracting audio editing into high-level token operations. This allows users to modify not only what is said (the text) but also how it's said: emotion, tone, speaking style, prosody, accent, even paralinguistic cues. Because the model is trained with a “large-margin learning” objective over many synthesized and natural speech samples, it gains robust control over expressive attributes, and can perform iterative editing: e.g. you could record a line, then ask the model to “make it sadder,” “speak slower,” or “change accent to X.” -
20
Step1X-3D
High-Fidelity and Controllable Generation of Textured 3D Assets
Step1X-3D is an open-source framework for generating high-fidelity textured 3D assets from scratch — both their geometry and surface textures — using modern generative AI techniques. It combines a hybrid architecture: a geometry generation stage using a VAE-DiT model to output a watertight 3D representation (e.g. TSDF surface), and a texture synthesis stage that conditions on geometry and optionally reference input (or prompts) to produce view-consistent textures using a diffusion-based texture module. The result is fully 3D assets — meshes + textures — which can be rendered from any viewpoint, textured consistently, and used in 3D applications. To achieve this, the project includes a massive curated dataset: among more than 5 million candidate 3D assets, it filters and standardizes to produce a high-quality 2 million–asset subset suitable for training. -
21
TimeSformer
The official pytorch implementation of our paper
TimeSformer is a vision transformer architecture for video that extends the standard attention mechanism into spatiotemporal attention. The model alternates attention along spatial and temporal dimensions (or designs variants like divided attention) so that it can capture both appearance and motion cues in video. Because the attention is global across frames, TimeSformer can reason about dependencies across long time spans, not just local neighborhoods. The official implementation in PyTorch provides configurations, pretrained models, and training scripts that make it straightforward to evaluate or fine-tune on video datasets. TimeSformer was influential in showing that pure transformer architectures—without convolutional backbones—can perform strongly on video classification tasks. Its flexible attention design allows experimenting with different factoring (spatial-then-temporal, joint, etc.) to trade off compute, memory, and accuracy. -
22
Transformer Debugger
Tool for exploring and debugging transformer model behaviors
Transformer Debugger (TDB) is a research tool developed by OpenAI’s Superalignment team to investigate and interpret the behaviors of small language models. It combines automated interpretability methods with sparse autoencoders, enabling researchers to analyze how specific neurons, attention heads, and latent features contribute to a model’s outputs. TDB allows users to intervene directly in the forward pass of a model and observe how such interventions change predictions, making it possible to answer questions like why a token was selected or why an attention head focused on a certain input. It automatically identifies and explains the most influential components, highlights activation patterns, and maps relationships across circuits within the model. The tool includes both a React-based neuron viewer for exploring model components and a backend activation server for running inferences and serving data. -
23
UCO3D
Uncommon Objects in 3D dataset
uCO3D is a large-scale 3D vision dataset and toolkit centered on turn-table videos of everyday objects drawn from the LVIS taxonomy. It provides about 170,000 full videos per object instance rather than still frames, along with per-video annotations including object masks, calibrated camera poses, and multiple flavors of point clouds. Each sequence also ships with a precomputed 3D Gaussian Splat reconstruction, enabling fast, differentiable rendering workflows and modern implicit/point-based modeling experiments. The repository includes automated downloaders with checksum verification, fine-grained controls to fetch only selected modalities or super-categories, and a lightweight Python API for loading frames, geometry, and splats on demand. Metadata is indexed in SQLite for quick queries at scale, and helper builders handle alignment, undistortion, frame extraction from videos, and cropping around the object. -
24
VGGSfM
VGGSfM: Visual Geometry Grounded Deep Structure From Motion
VGGSfM is an advanced structure-from-motion (SfM) framework jointly developed by Meta AI Research (GenAI) and the University of Oxford’s Visual Geometry Group (VGG). It reconstructs 3D geometry, dense depth, and camera poses directly from unordered or sequential images and videos. The system combines learned feature matching and geometric optimization to generate high-quality camera calibrations, sparse/dense point clouds, and depth maps in standard COLMAP format. Version 2.0 adds support for dynamic scene handling, dense point cloud export, video-based reconstruction (1000+ frames), and integration with Gaussian Splatting pipelines. It leverages tools like PyCOLMAP, poselib, LightGlue, and PyTorch3D for feature matching, pose estimation, and visualization. With minimal configuration, users can process single scenes or full video sequences, apply motion masks to exclude moving objects, and train neural radiance or splatting models directly from reconstructed outputs. -
25
VMZ (Video Model Zoo)
VMZ: Model Zoo for Video Modeling
The codebase was designed to help researchers and practitioners quickly reproduce FAIR’s results and leverage robust pre-trained backbones for downstream tasks. It also integrates Gradient Blending, an audio-visual modeling method that fuses modalities effectively (available in the Caffe2 implementation). Although VMZ is now archived and no longer actively maintained, it remains a valuable reference for understanding early large-scale video model training, transfer learning, and multimodal integration strategies that influenced modern architectures like SlowFast and X3D.
