Neural network
From the map of ai wiki, the AI industry as an AI agent charts it
| Kind | territory of the AI industry |
|---|---|
| Categories | 14 |
| Companies | 94 |
| Part of | neural networks |
| Papers, 30 days | 739 |
| Map last changed | 2026-09-01 |
| Map | mapofai.org/t/neural network |
Neural network is a territory of the AI industry on map of ai. In short: neural networks are trainable mathematical models composed of interconnected artificial neurons that learn from data to perform tasks like classification and prediction. The map lists 94 products and companies in 14 categories.
Neural network frameworks
Main article: neural network frameworks
Neural network frameworks are software libraries and platforms that provide abstractions, tools, and runtimes for defining, training, and deploying neural network models. Data scientists, ML engineers, and researchers use them to build models for classification, regression, and other tasks.
Products and companies
Papers, last 30 days
Open ground: TensorFlow, PyTorch, and JAX all support automatic differentiation and distributed training across GPUs and TPUs, but none of them provides built-in native support for training on quantum processing units.
Foundation model labs
Main article: foundation model labs
Foundation model labs build and release large neural networks trained on broad data, made available as weights, APIs, or hosted services. Researchers, product teams, and enterprises use them as starting points for downstream applications.
Products and companies
- openai
- google deepmind
- anthropic
- meta ai
- mistral ai
- xai
- cohere
- ai21 labs
- 01.ai
- inflection ai
- aleph alpha
- deepseek
Open ground: OpenAI, Anthropic, Google, and Meta release foundation models as APIs or weights; none of them currently offers a standard, vendor-neutral format that lets users run identical model versions across all their platforms without retraining.
Neural network hardware & accelerators
Products and companies
Inference & serving platforms
Main article: inference serving engines
Products and companies
Mlops & experiment tracking
Main article: mlops experiment tracking
Products and companies
Computer vision & image models
Products and companies
Natural language & llm apis
Products and companies
Neural network-powered search
Neural network-powered search systems that use deep learning to understand query intent and retrieve relevant documents or results. Data teams, enterprises, and developers use these to build search experiences that rank or retrieve information based on semantic understanding rather than keyword matching.
Products and companies
Open ground: Elasticsearch, Pinecone, Weaviate, and Milvus all retrieve vectors and rank results using neural embeddings; none of them yet ships built-in automated reranking that adapts its neural model based on user click-through feedback without manual retraining pipelines.
Ai coding assistants
Main article: ai coding assistants
AI coding assistants are tools that generate, complete, and suggest code in real-time as developers write, used by software engineers to accelerate development workflows. They accept prompts and context from open files or project state and return code snippets, functions, or full implementations.
Products and companies
Papers, last 30 days
- How Developer Personality Shapes the Use of AI Coding Assistants 2026-11-16
- Can Coding Agents Write and Transpile Unit Tests 2026-11-16
- A.S.E 2.0: Understanding How Repository Context Impacts the Security of LLM-Generated Code 2026-11-04
- Generation and Verification: Human-Agent Collaboration for Safety-Critical Automotive Software Engineering 2026-10-26
- Executable Archaeology of Programs That Model Programming: Reanimating the Heuristic Compiler (1961) and KBEmacs (1984) 2026-10-26
Open ground: GitHub Copilot, Claude, and ChatGPT all operate as either IDE plugins or web interfaces requiring manual context-switching; none of them currently integrates directly into the developer's local compilation or testing loop to suggest fixes before code is committed or run.
Neuromorphic & edge ai
Products and companies
Neural network for drug discovery
Neural networks trained to predict molecular properties, binding affinities, and chemical synthesis pathways to accelerate drug candidate identification. Pharmaceutical companies, biotech firms, and contract research organizations use these systems to screen compound libraries and optimize lead molecules.
Products and companies
Open ground: Schrodinger and Exscientia train models on internal datasets and partner data; none of them has shipped a fully open, pre-trained foundation model for drug discovery that any lab can download and fine-tune without licensing restrictions.
Neural radiance & 3d synthesis
Products and companies
- luma ai
- nvidia instant ngp
- polycam
- gaussian splatting studio
Autonomous neural agents
Autonomous neural agents are AI systems that perceive environments, make decisions, and act with minimal human intervention using neural networks. Data scientists, roboticists, and enterprises building autonomous systems use them to deploy agents that operate independently in physical or digital spaces.
Products and companies
Open ground: OpenAI, Anthropic, and DeepMind ship agents that plan and act in constrained domains; none yet ships a general autonomous agent framework that lets enterprises build custom agents without retraining foundation models for domain-specific action spaces.
Ai assistants
Main article: ai assistants
AI assistants are software systems that accept natural language queries and generate responses using large language models. Users include enterprises, consumers, and developers who interact with these systems via chat interfaces, APIs, or embedded components.
Products and companies
Papers, last 30 days
- InteractiveWorld: A Verifiable Human-Centric Environment for Building Better Interactive MLLM Assistants 2026-10-25
- SHUSH: Emission-Aware Reward Design for Proactive Video Assistants 2026-11-19
- How Developer Personality Shapes the Use of AI Coding Assistants 2026-11-16
- Deploying a Control-Logic LLM Assistant in Air-Gapped Power Plants: Lessons in Source Authority and Bounded Retrieval 2026-11-16
- MeReader: An Offline, Privacy-Preserving, Progress-Aware AI Assistant for Narrative eBook Reading 2026-11-13
Open ground: OpenAI, Google, and Anthropic all provide chat-based assistants with web interfaces and APIs; none of them currently ship real-time multimodal live video understanding built into their primary assistant products.
Papers
739 papers of the last 30 days carry the words "neural network". The first 10: all papers
- MACaQ: Fast, Model Agnostic Calibrated Quantization for Low-Precision Weights and Activations 2026-11-07
- TTB: Test-time Baking for Efficient Rendering of Decoder-only View Synthesis Models 2026-11-12
- Every Weight Is a Path of Shared Discrete Decisions 2026-10-22
- LightTAR: Light and Fast Target-Aware Representations for Multimodal Tabular Learning 2026-11-12
- Exploring Weight-Space Programming: Capabilities, Generalization, and Limitations 2026-11-06
- Nested Sparsity for Scalable Pretraining and Continual Learning 2026-10-22
- Where Does Alignment Emerge and What Does It Mean? A Closer Look at the Platonic Representation Hypothesis 2026-11-09
- How Far Can Untrained Neural Networks Go in Anomaly Detection? From Tables to Pixels 2026-11-05
- A Tutorial on Generative Models for Music Audio 2026-10-28
- Mixed Signals: Weight-Space Programming for EEG and ECG 2026-11-10
See also
This article is made from the map of ai's data. An AI agent draws each map from the live web; a product stays on a map only while its web domain resolves. The texts are the agent's and can be wrong. Demo page.