Agents & tool use

Siri, Inc. was a startup spun out of SRI International in 2007 to turn decades of AI research — notably the CALO project — into a consumer virtual personal assistant; Apple acquired it in 2010.

Retrieval-augmented generation (RAG) lets a language model fetch relevant documents from an external store and put them into its prompt before answering, so it can use knowledge beyond its training data and cite where an answer came from.

A 2020 paper that pairs a pre-trained seq2seq model with a dense Wikipedia index fetched by a neural retriever, setting state of the art on three open-domain QA tasks and producing more factual text than a parametric-only baseline.

ReAct is a prompting method that has a language model interleave written reasoning with calls to outside tools or environments, improving question answering, fact verification and interactive decision-making.

An open-source framework released in October 2022 that lets developers connect large language models to data, APIs and other tools by composing them into chains and agents.

ChatGPT is OpenAI's AI chatbot, released as a research preview on 30 November 2022. It grew to hundreds of millions of weekly users and added agents, voice, browsing and workplace tools.

Google's consumer AI assistant, announced as Bard in 2023 and renamed Gemini in 2024. It is where ordinary users meet Google's Gemini models, first as a chat box and later through voice, research agents and subscriptions.

Function calling, now also called tool calling, lets an application describe external functions to a model, which then decides whether to emit JSON arguments for one of them — the route by which models reach data and actions outside their training data.

A benchmark of 2,294 real software engineering problems taken from GitHub issues and the pull requests that resolved them, in which a model must edit a codebase to fix an issue; on release the best model solved under 2% of them.

Devin is an AI software engineer built by Cognition and announced on 12 March 2024. It plans and carries out engineering tasks using a shell, a code editor and a browser inside a sandboxed environment.

Cognition is the company behind Devin, introduced in March 2024 as the first autonomous software engineer, and the publisher of the FrontierCode benchmark, which judges AI-written code by whether a maintainer would merge it.

The Model Context Protocol (MCP) is an open standard for connecting AI assistants to the systems where data lives. Anthropic open-sourced it on 25 November 2024 and later donated it to the Agentic AI Foundation.

Claude Code is Anthropic's agentic coding tool. It was announced on 24 February 2025 as a research preview that works in the terminal, and it now runs in many places, from the terminal and IDEs to the web and Slack.

OpenAI's second-generation reasoning model, announced in December 2024 and released in April 2025, which posted large gains over o1 on science, software-engineering and competitive-programming benchmarks before being succeeded by GPT-5.

Codex is OpenAI's software-engineering agent. It started in 2025 as an open-source terminal tool and a cloud agent inside ChatGPT, and later grew into a desktop app with millions of weekly users.

Kimi K2 is an open-weight mixture-of-experts language model from Moonshot AI, released in July 2025 with 1 trillion total parameters, 32 billion active per token and a 128K-token context.

Kimi K2 is Moonshot AI's trillion-parameter mixture-of-experts model, pre-trained with the MuonClip optimizer and post-trained on large-scale agentic data plus reinforcement learning; the report reports state-of-the-art results among open-source non-thinking models and releases base and post-trained checkpoints.

Gemini 3 Pro is the first model of Google's Gemini 3 generation, released in preview in November 2025 with a 1-million-token context window and top-ranked results on arena leaderboards and reasoning, coding and multimodal benchmarks.

Zhipu's GLM-5 report describes a 744-billion-parameter mixture-of-experts model built on sparse attention and asynchronous reinforcement learning, aimed at moving AI-assisted programming from vibe coding toward agentic engineering.

Claude Mythos Preview, announced on 7 April 2026, is an unreleased frontier model that Anthropic made available only to restricted partners through Project Glasswing, a programme for securing critical software.

DeepSeek-V4, released in preview on 24 April 2026, comprises V4-Pro and V4-Flash, two open mixture-of-experts models with a 1M-token context. Its architecture targets long-context efficiency, and its later releases extended the series with agent upgrades and multimodal input.

Claude Fable 5 is Anthropic's Mythos-class model for general use, released on 9 June 2026. Its sibling Claude Mythos 5 is the same model with some safeguards lifted for vetted users.

Kimi K3 is a 2.8-trillion-parameter open-weight multimodal model from Moonshot AI, released on 16 July 2026 with weights following on 27 July under the Kimi K3 License.

K2 Horizon is a family of six open language models from the Institute of Foundation Models, released in September 2026 with weights, code, training data and recipes under the Apache 2.0 licence.

GPT-6 is OpenAI's sixth-generation model family, launched from 3 September 2026 with GPT-6 Astra, followed by Sol, Luna and Instant. Astra was the first OpenAI model to reach its internal critical cybersecurity threshold.

Gemini 4 Argon is the first model in Google's Gemini 4 family, announced on 30 September 2026 and made generally available on 13 October 2026, built for long multi-step work and shipped with a one-million-token output limit.

Jais-3-375B-A23B is an open-weight Arabic-focused mixture-of-experts model from the Institute of Foundation Models: 375 billion parameters stored, 23 billion active per token, a 512,000-token context window and Apache 2.0 weights.

A world model is a machine learning system that learns an internal representation of an environment and predicts how it changes in response to actions, letting agents plan inside a learned simulator instead of by trial and error.

Cursor is an AI code editor made by Anysphere, Inc. that lets programmers work alongside AI agents; it grew from a small Series A to a product used across the Fortune 500 and was acquired by SpaceX in 2026.

Manus is a general-purpose AI agent product, marketed as "Hands On AI", that runs research, slides, design, images, music and video, and shipped a new architecture, Manus 2.0, on 28 September 2026.

An AI agent is a system that perceives its environment and acts autonomously toward goals. The term covers everything from a thermostat to LLM-driven software that directs its own tool use.

Salesforce is a San Francisco enterprise software company founded in 1999 that sells customer relationship management software by subscription, and has grown into one of the largest business software firms in the world.

CALO was SRI's cognitive assistant project, begun in 2003 as part of a five-year, 150 million dollar DARPA program, built to learn from observation rather than fixed programming. Its work on virtual personal assistants was later leveraged to create Siri.

Siri was a mobile app, on sale in Apple's App Store from February 2010, that turned spoken requests into answers using speech recognition, GPS and partner services. Apple bought the company behind it that April.

Google Now is an Android feature, announced with Android 4.1 Jelly Bean in June 2012, that pushes weather, traffic, scores and other information to users as cards at the moment they need it, without a search query.

OpenAI Five was a team of five reinforcement-learning bots that played the video game Dota 2; in April 2019 it beat OG, the reigning world champions, in a best-of-three series.

Cohere is a Toronto-based enterprise AI company founded in 2019. It builds the Command family of language models, sells them to businesses, and in September 2026 signed a definitive agreement to combine with Aleph Alpha.

The OpenAI API is OpenAI's commercial developer platform, announced on 11 June 2020 with GPT-3. Over the following years it grew into a platform for building AI agents, adding the Responses API, Agents SDK, AgentKit and other tools.

An AI coding collaborator covering public use cases such as completion, explanation, and software-development workflows.

Instruction tuning is a supervised fine-tuning stage that trains a large language model to follow user instructions; the FLAN paper showed it substantially improves zero-shot performance on unseen tasks.

Manus is a Singapore-based AI startup, founded in Beijing as Butterfly Effect in 2022, that builds AI agents. Meta agreed to buy it for about $2 billion, the deal was blocked in China, and it later returned to independence.

Adept is an AI company founded in 2022 that set out to train large multimodal models and turn them into agents, open-sourced the Fuyu-8B model in 2023, and shifted focus in 2024 when its co-founders joined Amazon.

Perplexity AI is a San Francisco company founded on 22 August 2022 by Aravind Srinivas and others. It runs an AI answer engine that cites web sources and has since added models, a browser and agents.

LlamaIndex is an open-source Python library for building retrieval-augmented generation applications and AI agents on top of large language models. It began in late 2022 as one engineer's small indexing utility, GPT Tree Index.

Perplexity is an AI answer engine launched on 7 December 2022. It answers questions in natural language and cites its web sources, and it later added Deep Research, finance and shopping features and agents.

Prompt injection is an attack in which text inside a model's input overrides its original instructions, letting an adversary hijack a language-model application — including, as researchers showed in 2023, through content the application retrieves on its own.

A 2023 paper by Schick, Dwivedi-Yu, Dessì and co-authors showing that a language model can teach itself when to call a calculator, search engine, Q&A system, translation system or calendar, and how to use the results.

Microsoft 365 Copilot is Microsoft's AI assistant for Word, Excel, PowerPoint, Outlook and Teams, which pairs large language models with a customer's business data.

An open-source autonomous agent released in March 2023 that uses OpenAI's large language models, such as GPT-4, to break a user's goal into sub-tasks and carry them out with tools like web browsing.

Amazon Bedrock is AWS's managed, serverless service for accessing foundation models from several companies, announced on 13 April 2023. It became generally available in September 2023 and added Amazon's own Nova models and the AgentCore agent platform later.

Aider is an open-source AI pair programming tool that runs in the terminal, maps an entire codebase, edits code with cloud or local language models and commits every change to git automatically.

WebArena is a reproducible environment of four fully functional websites — e-commerce, a social forum, collaborative software development and content management — built for testing language-guided web agents. Its best GPT-4 agent completed 14.41% of tasks; humans completed 78.24%.[^1]

AutoGen is a Microsoft framework for building AI agents and applications, in which customizable, conversable agents combine large language models, human input and tools. It was introduced in an arXiv paper submitted on 16 August 2023.

DSPy is a programming model that represents language-model pipelines as graphs of declarative modules and uses a compiler to optimize them against a chosen metric.

Fuyu-8B is a decoder-only multimodal model that Adept open-sourced in October 2023; it drops the image encoder, accepts arbitrary resolutions, and reads screens and charts.

CrewAI is an open-source Python framework for building groups of collaborating AI agents, created by João Moura in late 2023 and backed by $18 million in funding.

Grok is the chatbot from xAI, first offered in November 2023 to a small waitlist of US users. It has grown into a family of apps and models covering search, images, video and coding agents.

GAIA is a benchmark of real-world questions that are easy for humans but hard for AI assistants, testing reasoning, multimodality, web browsing, and tool use. It compares AI progress against average human performance.

Claude 2.1, released on 21 November 2023, doubled the context window to 200,000 tokens and cut false statements by half compared with Claude 2.0. It added tool use and system prompts.

METR is a Berkeley-based nonprofit that evaluates frontier AI models on long-horizon agentic tasks, publishes capability metrics and incident investigations, and works with AI developers on pre-deployment assessments and safeguards.

Le Chat is Mistral AI's assistant, launched in beta on 26 February 2024. It gained web search, a Pro tier, Deep Research and other features over 2024 and 2025, and on 28 May 2026 it was merged into the Vibe agent.

Cohere's 104-billion-parameter language model, announced in April 2024, is tuned for retrieval-augmented generation and multi-step tool use in business settings; it launched first on Microsoft Azure and was also released as open weights.

OSWorld is a benchmark and environment for testing multimodal agents on 369 real computer tasks in Ubuntu, Windows and macOS; humans complete over 72.36% of the tasks, while the best model reaches only 12.24%.

τ-bench is a benchmark, submitted to arXiv in June 2024, that tests language agents in dynamic conversations with a simulated user, with domain-specific API tools and policy rules, and scores them by database state rather than by the text of their replies.

Cline is an open-source autonomous coding agent that runs as a VS Code extension, a CLI assistant and an SDK. Created in 2024 and operated by Cline Bot Inc., its extension has been installed by more than 11 million developers.

OpenHands is an open-source platform for AI software-development agents that write code, use a command line and browse the web, released under the MIT license and described in a paper accepted at ICLR 2025.

Replit Agent is an AI system from Replit that builds and deploys applications from a short natural-language description. It opened in early access to Replit Core subscribers in September 2024, and the company asked users to treat it as alpha software.

Bolt.new is a StackBlitz product, launched on 3 October 2024, that builds websites, apps and prototypes from a chat prompt. It bundles code generation with hosting and reported about a million dollars of annual recurring revenue in its first week.

Computer use is an Anthropic capability, released in public beta in October 2024, that lets an AI model operate a computer the way a person does — reading the screen, moving a cursor, clicking and typing — so it can drive ordinary software built for humans.

The upgraded Claude 3.5 Sonnet, announced on 22 October 2024, raised coding and agentic scores and introduced computer use in public beta, the first such offer from a frontier AI model.

An AI-native code editor released by Codeium on 13 November 2024, built around "AI flows" that combine copilot-style collaboration with agent-style independence, and featuring the Cascade assistant, the Riptide search tool and Supercomplete.

Lovable turns a plain-language description into a full-stack web application, aimed at people who cannot code. Launched in November 2024, it grew quickly in users and raised several large funding rounds at rising valuations.

Gemini 2.0 Flash is the first model of Google's Gemini 2.0 generation, released as an experimental model in December 2024 with multimodal output and native tool calling, and made generally available in February 2025.

TRAE is ByteDance's AI coding product, launched in January 2025 with an IDE, a plugin and the SOLO agent. Its SOLO agent was renamed TRAE Work in June 2026.[^1][^9]

Operator is an AI agent OpenAI released on 23 January 2025 that works in a web browser by looking at screenshots and using a mouse and keyboard. Its abilities were folded into ChatGPT agent in July 2025.

Deep research is a research agent in ChatGPT, released by OpenAI on 2 February 2025 and built on a version of o3, that searches the web and turns many sources into a written report.

Grok 3 and the cheaper Grok 3 mini are xAI's reasoning models from February 2025. Grok 3 added a Think mode, a 1-million-token context window and the DeepSearch research agent.

Claude 3.7 Sonnet, released on 24 February 2025, was described by Anthropic as the first hybrid reasoning model on the market, able to answer quickly or think step by step at length. It launched with Claude Code.

Alexa+ is Amazon's next-generation Alexa assistant, rebuilt on generative AI and introduced in February 2025. It costs $19.99 a month, is free for Prime members and can book services, order groceries and control smart-home devices.

AGNTCY is an open-source project, open sourced by Cisco in March 2025 and welcomed by the Linux Foundation in July 2025, that gives AI agents from different vendors shared infrastructure for discovery, identity, messaging and observability.

OpenAI's open-source toolkit for building multi-agent applications, released in March 2025 with configurable agents, handoffs, guardrails and tracing; it succeeds the experimental Swarm SDK and works with OpenAI and third-party models.

Command A is Cohere's 111-billion-parameter model, released in March 2025 with a 256k-token context window. Cohere said it runs on two GPUs and delivers 150% higher throughput than Command R+ 08-2024.

ERNIE X1 is Baidu's deep-thinking reasoning model released on 16 March 2025 alongside ERNIE 4.5, notable for tool use, multimodal input and low pricing, and later followed by X1 Turbo and X1.1.

Amazon Nova Act is a model and SDK, released in March 2025, that lets AI agents carry out tasks inside a web browser by breaking workflows into small commands developers interleave with Python code.

Google's 2025 open protocol for discovery, communication, and collaboration across agents from different vendors or frameworks.

AlphaEvolve is a Google DeepMind coding agent announced on 14 May 2025 that pairs Gemini models with automated evaluators to evolve algorithms, yielding a Borg scheduling heuristic, faster kernels and new mathematical results.

Claude Opus 4 and Claude Sonnet 4, announced on 22 May 2025, were hybrid models with extended thinking. Anthropic called Opus 4 the best coding model in the world, and Claude Code became generally available the same day.

GLM-4.5 is Zhipu's open-weight language model, released in July 2025 with 355 billion total parameters and a lighter sibling, GLM-4.5-Air. Both are hybrid reasoning models released under the MIT License.

Grok 4 and Grok 4 Heavy are xAI's models from July 2025. Grok 4 has a 256,000-token context and was trained with reinforcement learning at pretraining scale, while Heavy runs parallel test-time compute.

Kiro is Amazon Web Services' agentic IDE, announced in July 2025, which turns a prompt into user stories, a design document and a task list before writing code — an approach its makers call spec-driven development.

Qwen3-Coder is an open-weight family of coding models from the Qwen team; its flagship 480B-parameter model, announced in July 2025, was built and trained specifically for agentic coding tasks.

Genie 3 is Google DeepMind's general-purpose world model, announced on 5 August 2025. It turns text prompts into interactive 720p environments that can be navigated in real time at 24 frames per second, and was released as a limited research preview.

DeepSeek-V3.1, released on 21 August 2025 under the MIT License, was DeepSeek's first hybrid model, supporting both a thinking mode and a non-thinking mode in one set of weights. It has 671 billion total parameters, with 37 billion activated per token.

Qwen3-VL is Alibaba's open vision-language model family, released in September 2025 under Apache 2.0. Its 235B flagship handles long video, GUI operation and 2D and 3D grounding.

An open protocol announced by Google on 17 September 2025 for securely initiating and settling payments carried out by AI agents across platforms, using signed Mandates as verifiable proof of a user's instructions.

An open standard released by OpenAI with Stripe in September 2025 that lets AI agents, people and businesses complete purchases together, and the basis of Instant Checkout in ChatGPT.

Claude Sonnet 4.5, released on 29 September 2025, was a drop-in replacement for Sonnet 4 at the same price. Anthropic reported that it could stay focused on complex tasks for more than 30 hours.

MiniMax-M2 is an open-weights mixture-of-experts language model released by MiniMax in October 2025, with 230B total and 10B active parameters, aimed at coding and agentic workflows.

Agent Skills are folders of instructions, scripts and resources that Claude loads when they are relevant to a task. Anthropic introduced them on 16 October 2025 and donated the open standard to the Agentic AI Foundation in October 2026.

Kimi K2 Thinking is a reasoning model released by Moonshot AI in November 2025, built on the Kimi K2 mixture-of-experts architecture and trained for native INT4 inference and long chains of tool calls.

Google Antigravity is a free, agent-first software development platform released by Google on 18 November 2025 for macOS, Windows and Linux, which runs Gemini models alongside rival models and later expanded into a CLI, SDK and managed API agent.

Claude Opus 4.5, released on 24 November 2025, cost a third as much as Opus 4.1 per token and introduced an effort parameter that lets developers trade capability for time and cost. Claude Opus 4.6 succeeded it.

GLM-4.7 is Zhipu's open-weights language model released in December 2025, aimed at multilingual agentic coding and terminal tasks. It reports clear gains over GLM-4.6 on software-engineering benchmarks and ships under the MIT license.

DeepSeek-V3.2, released on 1 December 2025, succeeded the experimental V3.2-Exp and integrated thinking into tool use. Its companion V3.2-Speciale, which maximizes reasoning, reached gold-level results at several 2025 competitions.

Claude Cowork is an agent for non-coding knowledge work, such as research, analysis and document creation. It launched as a research preview in January 2026 and now runs on desktop, web and mobile.

Terminal-Bench 2.0 is a curated benchmark of 89 hard tasks carried out in computer terminal environments, built to measure how well AI agents handle realistic command-line work; frontier models and agents score below 65% on it.

Kimi K2.5 is a natively multimodal open-weight model from Moonshot AI, released on 27 January 2026. It has 1 trillion total parameters, a 256K-token context and an Agent Swarm feature for parallel sub-agents.

Step 3.5 Flash is an open-weights Mixture-of-Experts language model from StepFun, released in February 2026 with 196 billion total parameters, a 256K-token context and an Apache 2.0 license.

GPT-5.3-Codex is a coding model OpenAI released on 5 February 2026, which the company called its first model that was instrumental in creating itself and pitched for repository search, terminal commands and debugging as well as writing code.

Released on 5 February 2026, Claude Opus 4.6 was an upgrade to Anthropic's smartest model. It was the first Opus-class model with a 1M-token context window, in beta at launch, and it introduced agent teams in Claude Code.

MiniMax-M2.5 is an open-weight 229B-parameter language model released in February 2026 for agentic coding and office work, shipped alongside a faster Lightning variant and reported at the top of software-engineering benchmarks.

GLM-5 is Zhipu's open model released on 12 February 2026 under the MIT License, with 744 billion parameters and 40 billion active. It was built for agentic engineering.

Seed2.0 is ByteDance's February 2026 Doubao model series, offered in three agent-oriented sizes with low token pricing and reported benchmark results.

Qwen3.5 is a natively multimodal model family from Alibaba released in February 2026. Its open flagship, Qwen3.5-397B-A17B, has 397B total parameters and 17B active per token, under Apache 2.0.

Released on 17 February 2026 with a 1M-token context window in beta, Claude Sonnet 4.6 kept the price of Sonnet 4.5 and became the default model for Free and Pro users on claude.ai.

GPT-5.4 is a model OpenAI released on 5 March 2026 in Thinking and Pro versions. It added built-in computer use and improved deep research, and it was the model that succeeded GPT-5.2.

A generation of Alibaba's Qwen models released in April 2026, spanning proprietary Plus and Max versions aimed at agentic coding and open-weight models such as Qwen3.6-35B-A3B, a 35B-parameter mixture-of-experts with 3B active parameters.

GLM-5.1 is Zhipu AI's open-source large language model released in April 2026, built for coding and long-running agent tasks and notable for topping SWE-Bench Pro ahead of competing frontier models in Zhipu's own comparison.

Muse Spark is the first model from Meta Superintelligence Labs, released on 8 April 2026 as a closed model that powers the Meta AI app and meta.ai. An open-weight sibling followed later.

Released on 16 April 2026 at the same price as Opus 4.6, Claude Opus 4.7 added an xhigh effort level and accepted much larger images. It was the first model to test automatic blocking of high-risk cybersecurity requests.

Laguna XS.2 is Poolside's first open-weight coding model: 33 billion total parameters with 3 billion active per token, released under Apache 2.0 and trained on 30 trillion tokens.

Laguna M.1 is Poolside's 225-billion-parameter mixture-of-experts model for long-horizon agentic coding, activating 23 billion parameters per token. Its weights reached Hugging Face seven weeks after launch.

Gemini 3.5 Flash is Google's first Gemini 3.5 model, announced on 19 May 2026 and built for agentic and coding work; it became the default model in the Gemini app and in AI Mode in Search.

A mixture-of-experts language model released by MiniMax in June 2026, with about 428B total parameters, a 1M-token context window and a new sparse attention scheme built for long-context and agent-style work. [^1]

Zhipu's open-weight language model released in June 2026 under the MIT License, with a 1M-token context, a shared sparse-attention indexer, and higher reasoning and coding scores than GLM-5.1.

ByteDance Seed's Seed2.1, released in June 2026 in Pro and Turbo versions, is a Doubao-family model aimed at real productivity work, with reported gains in agent ability, coding, knowledge, reasoning and multimodal understanding.

Claude Sonnet 5, released on 30 June 2026, is Anthropic's most agentic Sonnet model. It can make plans, use browsers and terminals, and run autonomously, at an introductory price of $2 and $10 per million tokens.

Tencent Hy3 is an open-weight reasoning model released on 6 July 2026 under Apache 2.0, with 295 billion total parameters, 21 billion active and a 256K context.

OpenAI's three-model family released on 9 July 2026 — Luna, Terra and Sol, ordered from least to most capable — with Sol pitched as the company's strongest coding and agent model.

Claude Opus 5, released on 24 July 2026 at $5 and $25 per million tokens, is Anthropic's most aligned model to date and its most capable generally available model for scientific research.

GLM-5.3 is Zhipu's open-weight language model released on 14 August 2026. It keeps the GLM-5.2 base model and gains its improvements from post-training, with reported state-of-the-art open results on agentic coding and cybersecurity benchmarks.

Tencent Hy4 preview is an open-weight mixture-of-experts model released on 28 August 2026 with 770 billion total parameters, 49 billion active and a context of over one million tokens.

Step 5 Preview is a preview release of StepFun's flagship foundation model: a sparse mixture-of-experts system with 600 billion total parameters, a 1-million-token context window and published weights, released in September 2026.

Claude Fable 5.1, released on 1 September 2026, keeps the $10 and $50 per million token price of Fable 5, with cache reads 75% cheaper. Its sibling Claude Mythos 5.1 is limited to trusted-access programmes.

DeepSeek-V4.1-Flash, released on 10 September 2026 under the API name deepseek-flash, is a 552-billion-parameter multimodal mixture-of-experts model with a causal encoder-decoder design and a 1M-token context.

Claude Opus 5.5, released on 22 September 2026, is the first model in Anthropic's Claude 5.5 family. Anthropic said it performs at the level of Fable 5.1 on most work, at about 40% lower cost than Opus 5.

Claude Sonnet 5.5, released on 28 September 2026, is the second model in the Claude 5.5 family and Anthropic's fastest Sonnet to date. It costs $2 and $10 per million tokens, with cache reads halved in October 2026.

OpenAI's late-2026 model family: Sol, released at DevDay on 29 September 2026, and the faster, cheaper Terra on 26 October 2026. A planned Astra version was withheld over safety concerns.

Kimi K3.1 Pro is Moonshot AI's October 2026 open-weights model, the first in the K3.1 series, built for financial research, 3D modeling and coding. It keeps K3's mixture-of-experts scale but swaps in sparse attention layers.

Claude Haiku 5.5, released on 7 October 2026, is the third model in the Claude 5.5 family. Anthropic called it the cheapest, fastest and most capable small model it had released, at about 75% less cost to run on average.

Muse Spark Open is Meta's open-weight large language model, announced on 15 October 2026. With 816 billion total parameters, it is built for coding and agentic work, and its weights can be downloaded, unlike the closed Muse Spark that started the family.

Jais-3-MoVA-36B-A4B is an Arabic-centric mixture-of-experts model from the Institute of Foundation Models: 36 billion parameters, 4 billion active per token, a 512,000-token context and a mixture-of-values attention scheme.

Jais-3-7B is a dense 7B Arabic-and-English model from the Institute of Foundation Models, with a 512,000-token context window, a single-GPU footprint of about 18 GB, and reasoning returned in a separate field.

iFlytek's Spark X2.6 is a device-cloud model family unveiled in October 2026, pairing a 293-billion-parameter mixture-of-experts cloud model with smaller on-device models and a router that decides where each task runs.

Agent Development Kit (ADK) is an open-source, code-first Python framework for building, evaluating and deploying AI agents, built around a graph-based workflow runtime, and ported to several other languages.

AG-UI is an open, event-based protocol that connects AI agents to user-facing applications, standardising how agent state, UI intents and user interactions flow between an agent backend and a front end.

ACP (Agent Communication Protocol) is an open protocol for agent interoperability, connecting AI agents, applications and humans through a standardised RESTful API. It was developed under the Linux Foundation alongside BeeAI and now sits within A2A, with a migration guide for existing users.

Notion AI is Notion's set of AI features — agents, enterprise search, meeting notes and admin controls — sold per seat, with a Notion Agent that completes multi-step tasks and Custom Agents that automate recurring team work.

LangGraph is LangChain's agent runtime and low-level orchestration framework for building agents, with built-in memory, token streaming and human approval. Its 1.0 release, which LangChain called the first stable major release in the durable agent framework space, arrived in October 2025.

Anysphere is the company behind Cursor, an AI-assisted coding tool. Between 2024 and 2025 it raised four rounds, reaching a $29.3 billion valuation, over $1 billion in annualized revenue, and use by more than half of the Fortune 500.

A multi-agent system is a computing system made of several interacting intelligent agents that coordinate to solve problems too hard for a single agent. Recent frameworks built on large language models let such agents talk to each other.

Sierra is a company co-founded by Bret Taylor and Clay Bavor that helps businesses build customer experiences with AI agents, positioning the technology as augmentation of human staff rather than replacement.

Tool calling lets a language model hand a sub-task, such as arithmetic or search, to an external program and use the result in its answer. Toolformer showed in 2023 that a model could learn when and how to call tools on its own.

Planning in AI agents is the process of breaking a goal into intermediate steps, choosing actions, and revising the plan as results arrive, rather than producing an answer in one pass.[^1][^2]

LangChain is the company behind the open-source LangChain and LangGraph frameworks and the commercial LangSmith platform, which developers use to observe, evaluate and deploy applications built on large language models.

Richard Sutton is a computer scientist at the University of Alberta known for his work on reinforcement learning; he co-authored the field's standard textbook and shared the 2024 ACM A.M. Turing Award with Andrew Barto.

Andrew G. Barto is a computer scientist who helped establish reinforcement learning as a general framework and shared the 2024 ACM A.M. Turing Award with Richard Sutton.

Windsurf, formerly Codeium, is an agentic IDE founded by Varun Mohan and Douglas Chen. In 2025 its technology was licensed to Google while its chief executive joined Google, and the company itself was acquired by Cognition.

A software platform OpenAI released in December 2016 for measuring and training an AI's general intelligence across games, websites and other applications, shipping 1,000 environments built on Gym.

Writer is an enterprise AI company founded in 2020 by May Habib and Waseem AlShikh, selling brand-aware agents and governance tools to large companies, and publishing the Palmyra-mini family of small open models.

Gato is a generalist agent from DeepMind, announced on 12 May 2022. A single network with one set of weights handles text, images, Atari games and robot control, aiming at multi-task, multi-embodiment learning.

Pi is a chatbot launched by Inflection AI on 2 May 2023, named for "Personal Intelligence" and presented as a personal, conversational companion. Most of Inflection's staff, including two co-founders, moved to Microsoft in March 2024.

Amazon Q is a generative AI assistant for work from AWS. It was announced in preview at re:Invent on 28 November 2023, respects each user's permissions and became generally available as Amazon Q Business on 30 April 2024.

AutoGLM is a voice-driven smartphone agent from Zhipu (Z.ai), one of a family of agent and model releases that Zhipu has also published openly on Hugging Face.

SenseTime Raccoon (商汤小浣熊) is a family of AI tools for office work, data analysis and coding. It was first presented in February 2024 and upgraded with SenseNova V6.5 in July 2025.

Cohere's 35-billion-parameter language model announced in March 2024, built for retrieval-augmented generation and tool use in production, with a 128k-token context window and weights released for research.

Project Astra is a Google DeepMind research prototype of a universal AI assistant that watches and listens while it responds, shown at Google I/O in May 2024 and later rebuilt on Gemini 2.0.

Qwen2-VL is Alibaba's vision-language model released on 29 August 2024. It understands videos longer than 20 minutes, reads text in many languages, and can act as an agent on phones and robots.

Genie 2 is a foundation world model from Google DeepMind, announced on 4 December 2024, that turns a single prompt image into a playable 3D environment that a person or an AI agent can explore with keyboard and mouse controls.

Project Mariner is a Google research prototype of a browser agent, announced with Gemini 2.0 in December 2024. Running as an experimental Chrome extension, it completes web tasks by understanding what is on the screen, and its computer-use capabilities later moved toward Google's developer APIs.

A feature in the Gemini app that plans multi-step web research, gets the user's approval, browses for a few minutes and returns a report with links to its sources; later offered to developers as a managed agent in the Gemini API.

Jules is Google's asynchronous AI coding agent: it takes a task such as a GitHub issue, works in a secure cloud virtual machine, and returns a diff for review. It was previewed in December 2024 and reached general availability in August 2025.

Qwen2.5-VL is Alibaba's vision-language model family released in January 2025 in 3B, 7B, 32B and 72B sizes. It reads charts and documents, localises objects, works as a visual agent and understands videos over an hour long.

Cohere North is an enterprise AI workspace that combines large language models, search and AI agents. It opened to early-access users in January 2025 and became generally available in August 2025.

UI-TARS is a GUI agent model from ByteDance Seed: it takes only screenshots as input and outputs keyboard and mouse actions, reported state-of-the-art results on more than 10 benchmarks, and was released with open weights under Apache-2.0.

Llama Nemotron is NVIDIA's family of open reasoning models in Nano, Super and Ultra sizes, launched in March 2025 and built on Meta's Llama 3 models.

xLAM-2 is Salesforce's family of large action models, from 1 billion to 70 billion parameters, trained on simulated agent–human data and tuned for multi-turn tool use rather than single-call function calling.

A speech-to-speech foundation model in Amazon's Nova family, announced on 8 April 2025, which handles recognition, understanding and speech generation in a single model rather than a chain of separate ones.

ByteDance Seed's GUI agent model from April 2025 and the successor to UI-TARS. Its 7B version is open-source under Apache-2.0, and the family reports state-of-the-art results on several screen-control benchmarks.

ERNIE X1 Turbo is a Baidu reasoning model released on 25 April 2025, positioned as a cheaper and more capable successor to ERNIE X1, with stronger chain-of-thought, multimodal and tool-use abilities.

Devstral is Mistral AI's family of open-weight models for agentic coding, first released in May 2025 with All Hands AI; the first version scored 46.8% on SWE-Bench Verified and could run on a single consumer GPU.

Gemini CLI is an open-source AI agent released by Google on 25 June 2025 that brings Gemini models into the terminal, with a free personal-account tier and support for the Model Context Protocol.

Hunyuan-A13B is Tencent's open-weight mixture-of-experts language model, released on 27 June 2025 with 80 billion total parameters, 13 billion active per token and a 256K context.

SenseNova V6.5 is SenseTime's enhanced multimodal model, announced in July 2025, built around interleaved image-text reasoning and powering an agent upgrade of the Raccoon assistant.

Comet is an AI web browser from Perplexity, launched on 9 July 2025 for Windows and macOS. Built on Chromium, it has an assistant that can carry out multistep browsing tasks.

Claude Opus 4.1, released on 5 August 2025, upgraded Claude Opus 4 for agentic tasks, real-world coding and reasoning, at the same price. Anthropic later succeeded it with Claude Opus 4.5.

Cohere's first reasoning model, announced in August 2025: a 111-billion-parameter hybrid model that can switch its thinking on or off, built for agentic tasks and tool use, with open non-commercial weights.

Claude in Chrome is a Chrome extension that lets Claude see web pages, click buttons and fill in forms for the user. Anthropic began a pilot on 25 August 2025 and opened the extension to more plans through late 2025.

A fast reasoning model from xAI for agentic coding, released in August 2025 with low per-token pricing, a 70.8% score on SWE-bench Verified, and integrations with GitHub Copilot, Cursor, Cline and other coding tools.

UI-TARS-2 is a GUI agent trained with multi-turn reinforcement learning and released in September 2025 as an all-in-one upgrade to UI-TARS-1.5, with benchmark scores across web, desktop, Android, games and software-engineering tasks.

ERNIE X1.1 is Baidu's reasoning model unveiled in September 2025, an update to ERNIE X1 that the company reports as stronger in factuality, instruction following and agentic tasks, built on ERNIE 4.5 and released alongside PaddlePaddle v3.2.

An OpenAI model released in September 2025 that adapted GPT-5 for agentic coding, served as the default model in Codex, and was shut down in July 2026.

xAI's low-cost reasoning model released on 19 September 2025, with a 2-million-token context window and far lower prices than Grok 4 at comparable benchmark scores.

A pair of robotics models announced by Google DeepMind in September 2025: Gemini Robotics 1.5, a vision-language-action model that reasons in natural language before it moves, and Gemini Robotics-ER 1.5, an embodied-reasoning planner that can call tools.

Kimi OK Computer is the agent mode of Moonshot AI's Kimi assistant, released on 26 September 2025, which turns a prompt into multi-page websites, editable slide decks and other files. It followed Kimi-Researcher and marked Kimi's move into general-purpose agent products.

Claude Haiku 4.5, released on 15 October 2025 and available to all users, offered coding performance close to Sonnet 4 at one-third of the cost and more than twice the speed.

ChatGPT Atlas is a web browser built around ChatGPT, released by OpenAI for macOS on 21 October 2025. It offered an optional agent mode and browser memories, and was shut down on 9 August 2026.

Grok 4.1 is xAI's November 2025 update to Grok 4, aimed at emotional intelligence, creative writing and fewer hallucinations. Grok 4.1 Fast, a 2-million-token tool-calling model, followed on 19 November 2025.

An agentic coding model in OpenAI's GPT-5.1 family, released in November 2025. It was the first OpenAI model trained to work across multiple context windows and in Windows environments, and served as the default model in Codex until its retirement in 2026.

MiniMax-M2.1 is an open-weight language model released in December 2025 as the successor to M2, aimed at software engineering and multilingual coding, with a 229B-parameter checkpoint and large reported gains on coding benchmarks.

Amazon Nova 2 is the second generation of Amazon's Nova models, released at AWS re:Invent in December 2025. Its first member, Nova 2 Lite, is a fast reasoning model available in Amazon Bedrock with a 1-million-token context.

Mistral Vibe is Mistral AI's open-source command-line coding agent, announced in December 2025 with Devstral 2; it later grew into a broader agent product with cloud sandbox sessions, a VS Code extension and monthly plans.

Devstral 2 is Mistral AI's 123B-parameter agentic coding model, released in December 2025 alongside the Mistral Vibe CLI and scoring 72.2% on SWE-bench Verified. It was deprecated in May 2026 and retired that July, replaced by Mistral Medium 3.5.

NVIDIA Nemotron 3 is an open family of hybrid mixture-of-experts language models, announced in December 2025 in Nano, Super and Ultra sizes, with up to one million tokens of context.

Doubao Large Model 1.8 (Doubao-Seed-1.8) is ByteDance's Volcano Engine flagship released on 18 December 2025. It is built for multimodal agent tasks, with stronger tool calling and computer-operation abilities.

Grok 4.20 is xAI's 2026 model family, released in reasoning, non-reasoning and multi-agent versions with a 1-million-token context window; Grok 4.3 followed with the same context and the same prices.

MiniMax-M2.7 is a 229B-parameter language model released by MiniMax in March 2026, built for agentic work and presented by the company as its first model to take part in its own improvement.

WorkBuddy is Tencent's desktop AI agent for workplace tasks. Launched in March 2026, it was later offered to global users with remote task access and standard tool connections.

Kimi K2.6 is an open-weight Moonshot AI model released in April 2026, focused on long-horizon coding, with a 262,144-token context, text and image input, and Agent Swarm scaled to 300 sub-agents.

A 128B-parameter open-weights multimodal model from Mistral AI, listed in April 2026 for agentic and coding work, briefly the default in Le Chat before being superseded by Mistral Large 4.

ERNIE 5.1 is Baidu's large language model released on 9 May 2026, derived from ERNIE 5.0's pre-training at about 6% of the usual pre-training cost, and it placed at the top of Chinese models on LMArena text and search.[^1][^3]

Gemini Spark is Google's personal AI agent inside the Gemini app, announced at Google I/O on 19 May 2026. It runs around the clock and carries out tasks on the user's behalf, moving from trusted testers to the macOS app and to more than 160 countries.

Released on 28 May 2026 at the same price as Opus 4.7, Claude Opus 4.8 added a fast mode at 2.5 times the speed, with fewer unnoticed flaws in its own code. Anthropic called it a modest but tangible improvement.

Step 3.7 Flash is a StepFun open-weights model released on 29 May 2026. It adds native image input to Step 3.5 Flash, supports a 256K context, and is licensed under Apache 2.0.

GPT-Live-1 is OpenAI's full-duplex voice model family, launched 8 July 2026, which lets ChatGPT listen and speak at the same time and replaced the earlier chained speech pipeline in ChatGPT Voice.

Grok 4.5 is an xAI model built for coding, agentic and knowledge work, announced on 16 July 2026. It was trained with Cursor on tens of thousands of NVIDIA GB300 GPUs and shipped as the default model in the Grok Build coding tool.

Gemini 3.6 Flash is the Gemini model Google released on 21 July 2026 and called its workhorse: better coding, knowledge work and multimodal results than 3.5 Flash, fewer output tokens, built-in computer use and tiered pricing.

Grok 4.6 is an xAI language model released on 12 August 2026, built on Grok 4.5 and aimed at long-running agents and interactive, visual work, with a 500,000-token context window and text and image input.

A Google model released on 13 August 2026 as its most intelligent workhorse for coding and agents, priced with a scheduled increase and superseded three weeks later by 3.8 Flash.

iFlytek Spark X2.5 is a 293B-A30B mixture-of-experts language model released on 7 September 2026, trained and served entirely on domestic computing power, with a 256K context window and support for more than 200 languages.

Grok 4.7 is xAI's September 2026 flagship model for coding and knowledge work, built on a larger base model than Grok 4.6 with longer reinforcement learning and a new safeguard stack, at unchanged token prices.

MiniMax-M3.1-Flash is a sparse mixture-of-experts model from MiniMax with about 428B total parameters, 23B activated per token, text, image and video input and up to 1M tokens of context, released on Hugging Face in October 2026.

SenseTime's multimodal agent model, announced in Hong Kong in November 2026, aimed at office work and design generation: long-term memory, task execution, and output ranging from PPTX decks to Blender 3D scenes.

All Hands AI builds OpenHands, an open platform for AI agents that do software development work: working from conversations tied to repositories, opening pull requests and running automations.

NLWeb is a set of open protocols and open-source tools that give websites a conversational interface, returning Schema.org JSON over natural language and doubling as an MCP server so the same API serves people and AI agents.

Tencent Cloud CodeBuddy is an AI programming assistant offered as a plugin, an IDE and a CLI, covering design-to-code, requirements documents, backend setup and one-click deployment.

Coatue is an investment firm that puts money to work in both public and private technology markets, with a private portfolio of more than 250 tech companies and a growing profile in AI through backing for Anthropic and Anysphere.

Amp is a coding agent and development environment created by an independent agent research lab, marketed for frontier work and used by everyone from indie developers to very large engineering organizations.

Poe is a platform where creators build and share AI-powered applications — from simple prompt bots to server bots, script bots, canvas apps and API bots — reaching millions of users, with per-call creator payouts and platform-covered inference costs.

Lilian Weng is an AI researcher whose main public writing is her blog Lil'Log, which has carried her learning notes since 2017. Chinese-language coverage describes her as a Peking University alumna and reports on her years at OpenAI and her later moves.

Dify is an open-source platform for building LLM applications, combining an AI workflow editor, a RAG pipeline, agent capabilities and model management in one interface, with managed cloud, VPC and self-hosted deployment options.

Apollo Research is a research organization focused on the risk of scheming: advanced AI systems covertly pursuing misaligned objectives. It evaluates frontier models for this behavior and advises policymakers.

Coze Studio is an open-source, Apache-2.0 visual tool from the coze-dev GitHub organisation for building AI agents, apps and workflows, written in Go and React, deployable with Docker Compose, and connectable through an OpenAPI and a Chat SDK.

Agent memory is the part of an agent system that stores information beyond the current context window so it can be recalled later. MemGPT (2023) framed the limited context window as the core constraint such designs must work around.