---
id: platforms-consumer-developer
related:
  - platforms-enterprise
  - architecture-trends
  - consolidation-enterprise
key_findings:
  - "ChatGPT dominates at 900M WAU; Claude fastest enterprise growth at $14B ARR"
  - "Cursor at $2B ARR proves developer tools are a distinct category, not a feature of general assistants"
  - "Grok captured 17.8% US market share through X integration — distribution beats capability"
---

# Major AI Platforms Registry — Consumer & Developer
**As of March 22, 2026 | Compiled from primary sources**

---

## TABLE OF CONTENTS
1. [General Assistants](#general-assistants)
   - [ChatGPT (OpenAI)](#1-chatgpt-openai)
   - [Claude (Anthropic)](#2-claude-anthropic)
   - [Gemini (Google)](#3-gemini-google)
   - [Grok (xAI)](#4-grok-xai)
   - [Perplexity](#5-perplexity)
   - [DeepSeek](#6-deepseek)
   - [Meta AI](#7-meta-ai)
2. [Developer / Coding Tools](#developer--coding-tools)
   - [GitHub Copilot](#8-github-copilot)
   - [Cursor](#9-cursor-anysphere)
   - [Replit](#10-replit)
   - [Windsurf (Codeium)](#11-windsurf-codeium)
3. [Search / Research](#search--research)
   - [Perplexity (Search)](#12-perplexity-search-overlap)
   - [Google AI Overviews](#13-google-ai-overviews)
   - [You.com](#14-youcom)

---

## GENERAL ASSISTANTS

---

### 1. ChatGPT (OpenAI)

**Category:** General-purpose AI assistant / Agentic platform  
**Competitive Position:** Market leader by user volume; dominant platform for consumer AI

#### Current Status / Model Version
- **Active models (March 2026):** GPT-5.4 Thinking (launched March 5, 2026), GPT-5.4 mini (rolled out March 18, 2026), GPT-5.3 Instant (launched March 3, 2026)
- GPT-5.1 models deprecated March 11, 2026; all conversations auto-routed to GPT-5.3/5.4 equivalents
- GPT-5.4 Thinking merges reasoning, coding (GPT-5.3-Codex capabilities), and agentic workflows into a single frontier model; supports upfront thinking plans users can adjust mid-response
- Source: [OpenAI ChatGPT Release Notes](https://help.openai.com/en/articles/6825453-chatgpt-release-notes)
- Source: [releasebot.io ChatGPT March 2026 updates](https://releasebot.io/updates/openai/chatgpt)

#### Users & Revenue
- **900 million weekly active users** (announced February 27, 2026) — up 350% in 18 months
- **50 million paying consumer subscribers** (February 2026)
- **$25+ billion annualized revenue** (reported March 2026); $20B+ ARR reported January 2026 (3× YoY growth from $6B in 2024)
- 10%+ monthly growth confirmed by CEO Sam Altman, February 2026
- Source: [TechCrunch — ChatGPT reaches 900M weekly active users](https://techcrunch.com/2026/02/27/chatgpt-reaches-900m-weekly-active-users/)
- Source: [Reuters — OpenAI tops $25 billion annualized revenue](https://www.reuters.com/technology/openai-tops-25-billion-annualized-revenue-last-month-information-reports-2026-03-05/)
- Source: [OpenAI — A business that scales with the value of intelligence](https://openai.com/index/a-business-that-scales-with-the-value-of-intelligence/)

#### Pricing Tiers (March 2026)
| Plan | Price | Key Access |
|------|-------|------------|
| Free | $0 | GPT-5.4 mini Thinking (limited); ~10 messages/5 hrs |
| Go | $8/month | Unlimited GPT-5.3 Instant; no video/reasoning at full cap |
| Plus | $20/month | GPT-5.4 Thinking (capped), Sora (720p, limited), Agent Mode, Projects |
| Pro | $200/month | Unlimited GPT-5.4 (Instant, Thinking, Pro), Sora 1080p |
| Business | Custom | Enterprise controls, team admin |
| Enterprise | Custom | Custom deployment, data isolation |

- Source: [OpenAI ChatGPT Pricing](https://openai.com/business/chatgpt-pricing/)
- Source: [GlobalGPT ChatGPT Subscription Plans 2026](https://www.glbgpt.com/hub/chatgpt-subscription-plans-2026full-breakdown/)

#### Key Capabilities Shipped (Not Announced)
- **Agent Mode:** Autonomous web browsing, file management, form completion, multi-step task execution
- **Agentic Commerce Protocol (ACP):** Live since September 29, 2025 (co-developed with Stripe); "Instant Checkout" enables purchases directly in chat — Etsy live, Shopify (1M+ merchants including Glossier, SKIMS, Vuori) onboarding. Expanded to Free users February 16, 2026. 4% OpenAI transaction fee.
- **Operator History:** Persistent memory of completed agent tasks; audit trail of autonomous actions
- **Memory:** Cross-session user memory for personalization
- **Sora integration:** Video generation in-product for Plus/Pro tiers
- **ChatGPT Ads:** Launched January 2026 at ~$60 CPM; ad-supported free tier
- Source: [OpenAI — Buy it in ChatGPT / ACP](https://openai.com/index/buy-it-in-chatgpt/)
- Source: [AI Shopping Assistant Guide 2026 — Opascope](https://opascope.com/insights/ai-shopping-assistant-guide-2026-agentic-commerce-protocols/)

---

### 2. Claude (Anthropic)

**Category:** General-purpose AI assistant / Agentic knowledge worker platform  
**Competitive Position:** Premium enterprise AI and coding leader; highest-rated for coding (SWE-bench Lite #1)

#### Current Status / Model Version
- **Flagship:** Claude Opus 4.6 (launched February 5, 2026)
- Opus 4.6 improvements: State-of-the-art on Terminal-Bench 2.0 (agentic coding), Humanity's Last Exam (complex multidisciplinary reasoning), 1M token context window (beta), 128k output tokens, agent teams in Claude Code, adaptive thinking, compaction for long-running API tasks
- API pricing unchanged: $5/$25 per million input/output tokens (Opus 4.6 standard context)
- Anthropic valuation: **$380 billion** (CNBC, February 2026)
- Source: [Anthropic — Introducing Claude Opus 4.6](https://www.anthropic.com/news/claude-opus-4-6)
- Source: [TechCrunch — Anthropic releases Opus 4.6](https://techcrunch.com/2026/02/05/anthropic-releases-opus-4-6-with-new-agent-teams/)

#### SWE-bench Performance
- **Claude Opus 4.6: 62.7% on SWE-bench Lite** — #1 ranked model as of March 22, 2026 (top over GPT-5 at 54.3%, Gemini 2.5 Pro at 40.0%)
- Source: [SWE-bench Lite Leaderboard — Price Per Token](https://pricepertoken.com/leaderboards/benchmark/swe-bench-lite)

#### Users & Revenue
- No public MAU disclosed. Anthropic does not publish user figures.
- Valuation implies significant enterprise traction; Claude Code described as "indispensable" by engineering teams

#### Pricing Tiers (March 2026)
| Plan | Price | Key Access |
|------|-------|------------|
| Free | $0 | Limited Claude (below Opus-class) |
| Pro | $20/month ($17 annual) | 5× free usage, Google Workspace connectors, Cowork access |
| Max | $100/month | 5× Pro messages, Extended Thinking, Memory |
| Max (High) | $200/month | 20× Pro messages, Extended Thinking, Memory |
| Team | $25/user/month (annual) | Shared workspaces, team admin |
| Team Premium | $150/user/month | Claude Code full access |
| Enterprise | Custom | Custom deployment, data controls |

- Source: [Claude Pricing Page](https://claude.com/pricing)
- Source: [Claude AI Pricing 2026 — GlobalGPT](https://www.glbgpt.com/hub/claude-ai-pricing-2026-the-ultimate-guide-to-plans-api-costs-and-limits/)

#### Key Capabilities Shipped (Not Announced)
- **Cowork:** Desktop AI agent for knowledge workers — file access, web browsing, Gmail/Google Drive integration; enterprise connectors for DocuSign, FactSet, Google Drive live since February 24, 2026. Positioned as "indispensable for every knowledge worker."
- **Dispatch (launched March 18, 2026):** Remote control of Mac-based Cowork session from iPhone; currently Max subscribers, rolling to Pro users. Pairs via QR code. First-of-kind mobile-to-desktop AI orchestration.
- **MCP (Model Context Protocol):** Open standard for connecting AI to external tools; Claude supports MCP natively enabling developer ecosystem integrations
- **Claude Code:** Background coding agent; team assembly of agent squads working in parallel across codebases; PowerPoint native integration (research preview Feb 2026); Excel deep integration
- **Agent Teams:** Multiple Opus 4.6 agents collaborate on tasks (e.g., autonomously closed 13 GitHub issues across a 50-person org in a single day)
- Source: [MacStories — Hands-On with Claude Dispatch for Cowork](https://www.macstories.net/stories/hands-on-with-claude-dispatch-for-cowork/)
- Source: [Forbes — Claude Dispatch Controls Cowork With Phone](https://www.forbes.com/sites/ronschmelzer/2026/03/20/claude-dispatch-lets-you-control-claude-cowork-with-your-phone/)
- Source: [CNBC — Anthropic updates Claude Cowork](https://www.cnbc.com/2026/02/24/anthropic-claude-cowork-office-worker.html)

---

### 3. Gemini (Google)

**Category:** General-purpose AI assistant / Google ecosystem integration layer  
**Competitive Position:** #2 by MAU among standalone AI apps; deepest multimodal + search integration

#### Current Status / Model Version
- **Current flagship:** Gemini 3.1 Pro (launched February 19, 2026 in preview)
  - Most advanced model in Gemini 3 series; improved reasoning and multimodal capabilities; rolling out to API, Vertex AI, AI Studio, Gemini app, NotebookLM
  - Gemini 3 Pro released November 2025; 3.1 Pro is the rapid iteration
  - **Deep Think** version also updated (major update week before 3.1 Pro)
- **Current image models:** Nano Banana Pro (high-fidelity, available to AI Pro/Ultra subscribers); Nano Banana 2 launched February 26, 2026 as new default across all Gemini products
  - Nano Banana 2 = technically Gemini 3.1 Flash Image; 512px–4K resolution; 5-character consistency; faster than Pro
  - Nano Banana Pro remains available via three-dot menu for AI Pro/Ultra subscribers
- Source: [Google Blog — Gemini 3.1 Pro](https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-1-pro/)
- Source: [Google Blog — Nano Banana 2](https://blog.google/innovation-and-ai/technology/ai/nano-banana-2/)
- Source: [TechCrunch — Google launches Nano Banana 2](https://techcrunch.com/2026/02/26/google-launches-nano-banana-2-model-with-faster-image-generation/)

#### Users & Revenue
- **750 million+ monthly active users** on Gemini app (announced Q4 2025 earnings, February 4, 2026; up from 650M prior quarter)
- **~2 billion monthly users** for Gemini AI Overviews (via Google Search)
- Gemini API: ~2.4 million active developers (+118% YoY)
- Processes 10 billion+ tokens per minute via direct API
- Source: [TechCrunch — Google's Gemini app surpasses 750M MAU](https://techcrunch.com/2026/02/04/googles-gemini-app-has-surpassed-750m-monthly-active-users/)
- Source: [fatjoe — Google Gemini Stats 2026](https://fatjoe.com/blog/google-gemini-stats/)

#### Pricing Tiers (March 2026)
| Plan | Price | Key Access |
|------|-------|------------|
| Free | $0 | Gemini 3 Pro access (limited), Nano Banana 2 images |
| AI Pro | $19.99/month | Higher limits, Nano Banana Pro images, Veo 3.1 video |
| AI Ultra | Higher tier | Maximum limits, all models |
| Developer (AI Studio) | Free tier + pay-as-you-go | Gemini 3.1 Flash Lite Preview: $1.50/M output; Gemini 3.1 Pro: $2–$4/$12–$18/M tokens |

- Source: [Finout — Gemini Pricing 2026](https://www.finout.io/blog/gemini-pricing-in-2026)
- Source: [Google AI Developer API Pricing](https://ai.google.dev/gemini-api/docs/pricing)

#### Key Capabilities Shipped (Not Announced)
- **AI Studio:** Browser-based workspace for prototyping apps and agents; Gemini 3.1 Pro access, vibe-coding (natural language to app), multimodal input (text, images, audio, video), Veo 3.1 video generation (paid tier)
- **Workspace Integration:** Gemini embedded in Gmail, Docs, Sheets, Slides, Meet; Claude Pro also added Google Workspace connectors but Gemini is the native layer
- **NotebookLM:** Document-grounded research AI; Pro/Ultra users now access Gemini 3.1 Pro
- **Flow:** Video editing tool with Nano Banana 2 as default image model
- **SynthID + C2PA:** All Nano Banana images watermarked; interoperable with Adobe/Microsoft/OpenAI content credentials standard; used 20M+ times since November
- **Gemini CLI / Antigravity:** Agentic development platform; 3.1 Pro in preview on Android Studio
- Source: [Google DeepMind — Gemini 3.1 Pro Model Card](https://deepmind.google/models/model-cards/gemini-3-1-pro/)
- Source: [Website Builder Expert — Google AI Studio 2026](https://www.websitebuilderexpert.com/vibe-coding/google-ai-studio/)

---

### 4. Grok (xAI)

**Category:** General-purpose AI assistant / Real-time social intelligence platform  
**Competitive Position:** Fastest-growing US AI chatbot by market share; dominant in real-time/social intelligence; X/Twitter distribution moat

#### Current Status / Model Version
- **Current flagship:** Grok 4.20 — released February 17, 2026 (limited/internal beta), rolling to SuperGrok and X Premium+ users
  - Architecture: 4 AI agents collaborating; industry-leading agentic tool calling; lowest claimed hallucination rate; 2M token context window
  - Also in lineup: Grok 4 (July 9, 2025), Grok 4.1 Fast (speed/volume-optimized, 2M token context)
- **Grok Imagine:** Video generation tool; 10-second clips at 720p (February 2026); "Extend from Frame" for chaining clips up to 15s (March 2, 2026); built on Aurora autoregressive engine (110,000 NVIDIA GB200 GPUs); 1B+ monthly generations
- **DeepSearch mode:** Comprehensive web research with citations; operates like an in-context research agent
- Source: [xAI API Documentation — Models and Pricing](https://docs.x.ai/developers/models)
- Source: [NextBigFuture — xAI Launches Grok 4.20](https://www.nextbigfuture.com/2026/02/xai-launches-grok-4-20-and-it-has-4-ai-agents-collaborating.html)
- Source: [BASENOR — Grok Imagine March 2026 Update](https://www.basenor.com/blogs/news/grok-imagine-gets-a-major-update-whats-new-in-march-2026)

#### Users & Revenue
- **78+ million monthly active users** (AICPB research, early 2026)
- **60–62 million mobile app MAU** in January 2026 (+29% MoM from December)
- **314 million site visits** in January 2026 (Similarweb)
- U.S. market share: grew from 1.9% (January 2025) to **17.8%** (January 2026)
- 9% of users pay Premium; 25% MoM growth in paid users
- Source: [SeoProfy — Grok AI Usage Statistics 2026](https://seoprofy.com/blog/grok-ai-statistics/)
- Source: [fatjoe — Grok AI Stats 2026](https://fatjoe.com/blog/grok-ai-stats/)

#### Pricing Tiers (March 2026)
| Plan | Price | Grok Access |
|------|-------|------------|
| X Basic | $3/month | None |
| X Premium | $8/month | Standard Grok access (rate-limited) |
| X Premium+ / SuperGrok | $40/month | Full Grok 4.20 access; highest limits; ad-free X |
| API — Grok 4.1 Fast | $0.20/$0.50 per 1M in/out | 2M context; best speed/cost |
| API — Grok 4 | $3.00/$15.00 per 1M in/out | 256K context; complex reasoning |

- Source: [Data Studios — Grok Pricing and Subscriptions](https://www.datastudios.org/post/grok-pricing-and-subscriptions-explained-x-premium-and-premium-access-usage-tiers-and-feature-av)
- Source: [AI Free API — xAI Grok API Pricing 2026](https://aifreeapi.com/en/posts/xai-grok-api-pricing)

#### Key Capabilities Shipped (Not Announced)
- **Native X/Twitter Integration:** Real-time access to public X posts; semantic search of X content; compose-for-X content generation; X-integrated surface in left nav (desktop) and bottom menu (mobile)
- **Real-time live search:** Live web + X post search combined; developer-controlled via `search_parameters` API field; priced per source used
- **DeepSearch:** Multi-step research with live retrieval and citation
- **Big Brain mode:** Extended compute for complex analytical tasks
- **Grok Imagine:** Image and video generation; Extend from Frame (March 2026); Folders for content organization

---

### 5. Perplexity

**Category:** Search-native AI assistant / Multi-model orchestration platform / Enterprise research  
**Competitive Position:** "Answer engine" leader; pivoting from broad consumer to high-value enterprise research; anti-scale boutique strategy

#### Current Status
- **Perplexity Computer:** Launched February 25, 2026; orchestrates **19 AI models** for research, code, design, and deployment. Runs Opus 4.6 as core reasoning engine; routes subtasks to specialized models (Gemini for deep research/sub-agents, Nano Banana for images, Veo 3.1 for video, Grok for speed, ChatGPT 5.2 for long-context recall)
- **Perplexity Computer for Enterprise:** Announced at "Ask 2026" developer conference (March 11, 2026 in San Francisco); includes Slack integration; multi-model agent for enterprise workflows
- **Draco benchmark:** Proprietary complex research benchmark released by Perplexity; Perplexity deep research outperforms Gemini on it
- **Comet browser:** iOS launch announced for April 2026
- Abandoned advertising business in late 2025 (said it "undermined users' trust in accuracy")
- Source: [Perplexity — Introducing Perplexity Computer](https://www.perplexity.ai/hub/blog/introducing-perplexity-computer)
- Source: [TechCrunch — Perplexity's new Computer](https://techcrunch.com/2026/02/27/perplexitys-new-computer-is-another-bet-that-users-need-many-ai-models/)
- Source: [VentureBeat — Perplexity takes Computer to enterprise](https://venturebeat.com/technology/perplexity-takes-its-computer-ai-agent-into-the-enterprise-taking-aim-at)

#### Users & Revenue
- **~30 million monthly active users** (April 2025 figure; most recent disclosed); ~780 million monthly queries
- **ARR ~$200 million** as of October 2025 (per reporting); $150M+ by mid-2025
- **$20 billion valuation** (raised $1.5 billion total)
- Internal ARR target: **$656 million by end of 2026** (Sacra estimates, Perplexity internal projections)
- Executives explicitly not tracking MAU: "We're not actually on a mission to get as many users as possible" — prioritizing revenue and "GDP-moving decisions"
- Source: [Business Insider — Perplexity shifts to subscriptions](https://www.businessinsider.com/perplexity-shifts-to-subscriptions-business-growth-2026-2)
- Source: [TechHQ — Perplexity enterprise AI](https://techhq.com/news/perplexity-enterprise-ai-computer-trust-ciso/)
- Source: [Sacra — How Perplexity hits $656M ARR](https://sacra.com/research/how-perplexity-hits-656m-arr/)

#### Pricing Tiers (March 2026)
| Plan | Price | Key Access |
|------|-------|------------|
| Free | $0 | Limited daily queries; basic models; citations |
| Pro | $20/month ($200/year) | Unlimited Pro queries; 20 deep research/day; GPT-5, Claude 4.5; file uploads |
| Max | $200/month | Perplexity Computer; unlimited labs; advanced models (o3-pro, Opus 4.1 Thinking); 15 HQ videos/month |
| Enterprise Pro | $40/user/month ($400/year) | 500 research queries/day; SSO; SOC 2; Slack/Notion integration |
| Enterprise Max | $325/user/month ($3,250/year) | Unlimited queries; o3-pro; Computer for Enterprise |

- Source: [Screenapp — Perplexity Pricing 2026](https://screenapp.io/blog/perplexity-pricing)
- Source: [Finout — Perplexity Pricing 2026](https://www.finout.io/blog/perplexity-pricing-in-2026)
- Source: [Perplexity Enterprise Pricing](https://www.perplexity.ai/enterprise/pricing)

---

### 6. DeepSeek

**Category:** Open-weight AI model provider / Developer API / China AI positioning  
**Competitive Position:** Price-performance leader; open-weight strategy driving global developer adoption; structural cost advantage over US competitors

#### Current Status / Model Version
- **Current deployed models (March 2026):**
  - **DeepSeek V3.2:** Unified chat + reasoning model at same price; replaces V3 and R1 in API; released December 2025; 128K context; SWE-bench Lite: ~14% (V3.1 level, full V3.2 not yet rated)
  - **DeepSeek R1:** Reasoning-specialized; still available
- **DeepSeek V4 status:** As of March 22, 2026, **NOT YET RELEASED** publicly. Chinese tech media reported a website model update on March 9 (some calling it "V4 Lite") — unconfirmed. Expected in April per Reuters. Community predicted March window has passed.
- DeepSeek parent: Hangzhou-based hedge fund controlled by Liang Wenfeng; research-first, not VC-funded; no external investor pressure
- Source: [Reuters — A year after DeepSeek shock](https://www.reuters.com/world/china/year-deepseek-shock-get-set-flurry-low-cost-chinese-ai-models-2026-02-12/)
- Source: [evolink.ai — DeepSeek V4 Release Date tracker](https://evolink.ai/blog/deepseek-v4-release-window-prep)
- Source: [TLDL — DeepSeek API Pricing March 2026](https://www.tldl.io/resources/deepseek-api-pricing)

#### Users & Revenue
- **~97 million monthly active users** (April 2025 peak; #1 most downloaded app in App Store in 156+ countries in January 2025)
- **22 million+ daily active users** (January 2025 spike)
- No revenue disclosed (research-first structure; not commercially oriented)
- Source: [Backlinko — DeepSeek AI Usage Stats 2026](https://backlinko.com/deepseek-stats)

#### Pricing (API — March 2026)
| Model | Input /1M tokens | Output /1M tokens | Cache Hit |
|-------|-----------------|-------------------|-----------|
| DeepSeek V3.2 (Chat) | $0.28 | $0.42 | $0.028 |
| DeepSeek V3.2 (Reasoner) | $0.28 | $0.42 | $0.028 |

- Comparison: GPT-5 costs ~$1.25/$10/M tokens; Claude Opus 4.6 costs $5/$25/M tokens
- Open weights downloadable; self-hosting eliminates API costs entirely
- Consumer app: Free
- Source: [DeepSeek API Docs — Models & Pricing](https://api-docs.deepseek.com/quick_start/pricing)
- Source: [devtk.ai — DeepSeek API Pricing Guide 2026](https://devtk.ai/en/blog/deepseek-api-pricing-guide-2026/)

#### Key Capabilities Shipped
- **Open-weight release:** V3 and R1 released under Apache 2.0 (usable commercially); self-hostable via Ollama, LM Studio
- **Benchmark performance:** DeepSeek V3.2 Speciale: 96.0% AIME pass rate (vs GPT-5-High 94.6%); #2 on Artificial Analysis Intelligence Index among open-weight models (per Recode China AI)
- **OpenAI-compatible API:** Zero migration friction for developers switching from OpenAI
- **China open-source positioning:** Per MIT research, China's share of global open AI model downloads exceeds the US; DeepSeek + Qwen + Kimi driving this

---

### 7. Meta AI

**Category:** Embedded AI assistant / Open-source model provider  
**Competitive Position:** Largest user base globally (via social platform distribution); most downloaded open-weight model family (Llama)

#### Current Status / Model Version
- **Production model:** Llama 4 (Scout + Maverick released April 5, 2025; Behemoth still training)
  - Llama 4 Scout: 10M token context window (industry-leading); lightweight/fast
  - Llama 4 Maverick: General assistant + image understanding; 1M token context; available in 40 countries in WhatsApp/Messenger/Instagram/Facebook
  - Llama 4 Behemoth: Still in training as of March 2026
- Meta AI standalone app: Launched April 29, 2025; voice-first design; built on Llama 4
- Source: [Meta — The Llama 4 Herd](https://ai.meta.com/blog/llama-4-multimodal-intelligence/)
- Source: [Meta Newsroom — Introducing Meta AI App](https://about.fb.com/news/2025/04/introducing-meta-ai-app-new-way-access-ai-assistant/)
- Source: [TechCrunch — Meta releases Llama 4](https://techcrunch.com/2025/04/05/meta-releases-llama-4-a-new-crop-of-flagship-ai-models/)

#### Users & Revenue
- **1 billion+ monthly active users** across WhatsApp, Instagram, Facebook, Messenger (confirmed by Zuckerberg, Q1 2025 earnings; confirmed again October 2025)
- Fastest AI platform to 1 billion MAU in history
- 40 million daily active users
- 4 million businesses using Meta's generative AI ad tools
- Meta AI free; no separate subscription; revenue via advertising
- Source: [Meta Facebook/Zuckerberg — 1 Billion MAUs](https://www.facebook.com/yahoofinance/videos/zuckerberg-meta-ai-has-more-than-a-billion-maus/1028832076005579/)
- Source: [aifundingtracker.com — Meta AI News 2026](https://aifundingtracker.com/meta-ai-news/)

#### Pricing
| Tier | Cost | Notes |
|------|------|-------|
| Meta AI consumer | Free | Embedded in WhatsApp, Instagram, Messenger, Facebook; standalone app |
| Llama 4 Maverick API | $0.15/$0.60 per 1M in/out | Via third-party providers (e.g., Together AI, Groq) |
| Llama 4 Scout API | $0.08/$0.30 per 1M in/out | Lightest/cheapest Llama 4 |

- Source: [Price Per Token — Llama 4 Maverick API Pricing](https://pricepertoken.com/pricing-page/model/meta-llama-llama-4-maverick)

#### Key Capabilities Shipped
- **Native multimodality:** Llama 4 uses early fusion to unify text + vision tokens
- **10M token context (Scout):** Longest context window of any released open-weight model
- **WhatsApp/Instagram/Messenger integration:** Conversational AI directly in chat threads; image generation; web search
- **Open-source strategy:** Llama 4 weights available on Llama.com and Hugging Face; downloadable for commercial use (with license terms)
- **Voice-first app:** Standalone Meta AI app emphasizes voice conversations; more personal/contextual than chat

---

## DEVELOPER / CODING TOOLS

---

### 8. GitHub Copilot

**Category:** AI coding assistant / Enterprise developer tool  
**Competitive Position:** Market share leader (42%) backed by Microsoft/GitHub distribution; strongest enterprise penetration

#### Current Status
- Full multi-file agent mode deployed; Workspace agent for PR-level changes
- **SWE-bench score: 46.3–56%** depending on source and test configuration (Morphllm comparison, March 4, 2026 puts Copilot at 56% on SWE-bench tasks vs. Cursor at 52%)
- Model picker supports multiple models (including Claude, GPT-5 variants)
- **20 million cumulative users** as of July 2025 (1.3 million paid); 90% of Fortune 100 companies adopted
- Generates **46% of code** written by users on average (Java developers: 61%)
- Microsoft CEO confirmed Copilot now represents a larger business than GitHub itself was at acquisition in 2018
- Source: [Quantumrun — GitHub Copilot Statistics 2026](https://www.quantumrun.com/consulting/github-copilot-statistics/)
- Source: [Morphllm — Cursor vs Copilot 2026](https://morphllm.com/comparisons/cursor-vs-copilot)

#### Market Share
- **42% of AI coding assistant market** (market size: $7.37 billion in 2025; projected $30.1B by 2032 at 27.1% CAGR)
- Cursor has 18%; Amazon Q Developer and others hold the rest
- Source: [Quantumrun — GitHub Copilot Statistics 2026](https://www.quantumrun.com/consulting/github-copilot-statistics/)

#### Pricing Tiers (2026)
| Plan | Price | Notes |
|------|-------|-------|
| Free | $0 | 2,000 completions/month; 50 chat requests/month |
| Pro | $10/month ($100/year) | 300 chat requests/month |
| Pro+ | $39/month ($390/year) | 1,500 chat requests/month; priority access |
| Business | $19/user/month | Team admin; 300 requests/user |
| Enterprise | $39/user/month | 1,000 requests/user; policy management |

- Source: [UserJot — GitHub Copilot Pricing 2026](https://userjot.com/blog/github-copilot-pricing-guide-2025)
- Source: [GitHub Copilot Plans](https://github.com/features/copilot/plans)

---

### 9. Cursor (Anysphere)

**Category:** AI-native IDE / Agentic coding platform  
**Competitive Position:** Fastest-growing coding tool; highest valuation of any developer tool startup; strong enterprise momentum

#### Current Status / ARR / Valuation
- **$2 billion+ annualized revenue** (reported March 2, 2026 by TechCrunch/Bloomberg) — doubled in 3 months
- **~$50 billion valuation** in active funding talks (Bloomberg, March 12, 2026) — nearly double the $27B valuation from fall 2025
- **60% of revenue from enterprise** (corporate buyers)
- Source: [TechCrunch — Cursor surpasses $2B annualized revenue](https://techcrunch.com/2026/03/02/cursor-has-reportedly-surpassed-2b-in-annualized-revenue/)
- Source: [Bloomberg — Cursor in talks for ~$50B valuation](https://www.bloomberg.com/news/articles/2026-03-12/ai-coding-startup-cursor-in-talks-for-about-50-billion-valuation/)

#### Key Capabilities Shipped
- **Automations (launched March 5, 2026):** Event-triggered agent workflows — launch agents from Slack messages, new code additions, or timers; Bugbot (security audits, bug review), PagerDuty incident response, weekly Slack changelog summaries
- **Cloud VM agents:** 35% of Cursor's own internal PRs now created by autonomous agents in cloud VMs
- **Tab model:** Specialized autocomplete model optimized for prediction speed and precision
- **Multi-agent parallel execution:** Developers can run multiple agents simultaneously on independent tasks
- **MCP integration:** Connects agents to external tools via MCP connections
- Source: [TechCrunch — Cursor rolls out Automations agentic coding](https://techcrunch.com/2026/03/05/cursor-is-rolling-out-a-new-system-for-agentic-coding/)
- Source: [Cursor Blog — The third era of AI software development](https://cursor.com/blog/third-era)

#### SWE-bench / Benchmarks
- **SWE-bench Verified:** 51.7% (Tech Insider, March 2026) vs. Copilot 46.3%
- Cursor 52% SWE-bench tasks; 30% faster to complete than Copilot (Morphllm, March 4, 2026)
- Source: [Tech Insider — GitHub Copilot vs Cursor 2026](https://tech-insider.org/github-copilot-vs-cursor-2026/)

#### Pricing
| Plan | Price | Notes |
|------|-------|-------|
| Hobby | $0 | Limited agent use |
| Pro | $20/month | Full agent access; model selection |
| Business | $40/user/month | Team management; enterprise controls |

---

### 10. Replit

**Category:** Cloud IDE + AI agent platform / No-code/low-code builder  
**Competitive Position:** Dominant cloud-based AI development environment; targets non-engineers and vibe coders; enterprise collaboration focus

#### Current Status
- **Agent 4 (launched February 24, 2026):** Built around four pillars: Design Freely (visual infinite canvas → live code), Build Together (multi-user real-time collaboration), Ship Anything (full-stack app deployment), Move Faster (parallel agents on simultaneous tasks)
- Kanban-based multi-agent task management for enterprise
- Figma/Lovable design import directly into Replit AI
- Microsoft Azure partnership for cloud deployment
- Rokt built 135 internal applications in 24 hours with Replit Agent (enterprise case study)
- Source: [Replit — Introducing Agent 4](https://replit.com/agent4)

#### Pricing Tiers (March 2026)
| Plan | Price | Key Features |
|------|-------|--------------|
| Starter | Free | Daily agent credits; 1 published app; limited intelligence |
| Core | $25/month ($20/month annual) | $25 monthly credits; 5 collaborators; unlimited workspaces; autonomous long builds |
| Pro | $100/month ($95/month annual) | $100 monthly credits; 15 collaborators; 50 viewers; private deployments; premium models |
| Enterprise | Custom | Custom seats; SSO/SAML; VPC peering; static IPs; region selection; data warehouses |

- Source: [Replit Pricing](https://replit.com/pricing)
- Source: [Superblocks — Replit Pricing Breakdown 2026](https://www.superblocks.com/blog/replit-pricing)

#### Key Capabilities Shipped
- **Visual design canvas:** Explore and tweak designs visually; apply directly to app code
- **Parallel agent execution:** Multiple agents tackle auth, database, back-end, and front-end simultaneously; progress always visible
- **Browser-based deployment:** One-click publish to public URL; no local setup required
- **Effort-based pricing:** Credits consumed by agent intelligence level and task complexity

---

### 11. Windsurf (Codeium)

**Category:** AI-native IDE + coding agent  
**Competitive Position:** #3 in AI coding IDE; value alternative to Cursor; strongest agentic autonomy; lower price point

#### Current Status / Valuation
- Codeium raised $150M Series C at **$1.25 billion valuation** (August 2024, led by General Catalyst)
- In talks for new round at **$2.85 billion valuation** (February 2025, led by Kleiner Perkins)
- No 2026 valuation update found; most recent confirmed: $2.85B discussions in early 2025
- $40M ARR at time of 2025 fundraise discussions
- Source: [Windsurf — Series C Announcement](https://windsurf.com/blog/series-c-annoucement)
- Source: [TechCrunch — Codeium in talks to raise at ~$3B valuation](https://techcrunch.com/2025/02/19/ai-coding-startup-codeium-in-talks-to-raise-at-an-almost-3b-valuation-sources-say/)

#### Key Capabilities Shipped (Cascade Flow)
- **Cascade:** Core agentic system; Write mode (multi-file plan + execute) and Chat mode; tracks all user actions (edits, commands, clipboard, terminal output) to infer intent and stay in flow — "Flow Awareness"
- **Proprietary Fast Context:** Indexes entire codebase using embeddings; automatic context retrieval without manual file tagging — advantage over Cursor's @-mention system for large repos
- **Cascade beyond editor:** Integrated browser (send live preview context to Cascade); terminal auto-fix of lint errors (~60% success on common issues); web search within agent loop
- **Model Flexibility:** SWE-1.5 proprietary model family (speed-optimized for coding) + Claude + GPT model routing by task complexity
- **Teams Analytics:** Dashboard showing % new code written by Cascade, Tab completions, memory usage
- Cascade is more autonomous than Cursor's Agent (makes more decisions without asking); Cursor stays more collaborative
- Source: [Windsurf — Cascade](https://windsurf.com/cascade)
- Source: [pinklime.io — Windsurf (Codeium) Review 2026](https://pinklime.io/blog/windsurf-codeium-review-2026)
- Source: [nxcode.io — Windsurf AI Review 2026](https://www.nxcode.io/resources/news/windsurf-ai-review-2026-best-ide-for-beginners)

#### Pricing (2026)
| Plan | Price | Notes |
|------|-------|-------|
| Free | $0 | Limited Cascade uses |
| Pro | $15/month | Full Cascade access; model selection |
| Teams | $30/user/month | Analytics dashboard; shared settings |
| Enterprise | Custom | Custom deployment |

---

## SEARCH / RESEARCH

---

### 12. Perplexity (Search Overlap)

*(Full entry above in General Assistants — see Section 5)*

**Search-specific notes:**
- Processes ~780 million monthly search queries (April 2025 data; most recent)
- Core product is a search-native answer engine: every response cites sources, always-on web access
- Deep Research: Multi-step research agent producing comprehensive reports (20 research queries/day on Pro; unlimited on Enterprise Max)
- Introduced Draco benchmark for measuring complex research quality
- **Model Council:** Query multiple AI models simultaneously and compare outputs
- Sonar API (developer search API): $5 per 1,000 requests for raw web results; Sonar Deep Research: $2/$8/M tokens + $5/1K queries
- Source: [Perplexity — Introducing Perplexity Computer](https://www.perplexity.ai/hub/blog/introducing-perplexity-computer)

---

### 13. Google AI Overviews

**Category:** AI-native search layer (embedded in Google Search)  
**Competitive Position:** Largest-scale AI search deployment on earth; directly embedded in 2 billion monthly active search users' experience

#### Current Status
- AI Overviews launched widely May 2024; globally deployed across 141+ countries by early 2026
- **AI Mode:** Topped **75 million daily users** (as of mid-March 2026 per almcorp.com)
- AI Overviews now appear on **21% of all Google searches** (LinkedIn/Google data, March 2026)
- AI Overviews coverage **grew 58% year-over-year** across 9 industries (health, finance, tech, retail, travel, legal, education, real estate, food)
- Health queries: 82% now trigger an AI Overview before any organic result
- Source: [almcorp.com — Digital Marketing News March 11–20, 2026](https://almcorp.com/blog/digital-marketing-news-march-11-20-2026/)
- Source: [Emarketed — Google AI Overviews Grew 58%](https://emarketed.com/aeo/google-ai-overviews-surge-agency-guide-2026/)

#### Users & Scale
- ~2 billion monthly users via Google Search AI Overviews (accessed through standard Google search)
- Nano Banana 2 became the default image model in AI Mode across 141 countries (February 26, 2026)
- Google Lens integration; Gemini 3.1 Pro powers the underlying reasoning
- Source: [fatjoe — Google Gemini Stats 2026](https://fatjoe.com/blog/google-gemini-stats/)
- Source: [searcheseverywhere.com — Google AI Overviews 2026](https://www.searcheseverywhere.com/blog/google-ai-overviews-in-2026-search-data)

#### Impact on Search Behavior
- When AI Overview appears, organic CTR drops to **8%** (vs 15% without); ~47% traffic loss for top-ranking results
- Organic CTR declines of **up to 61%** on informational queries (industry study)
- AI-driven SERP features appear in **30–45%** of informational searches (Semrush data, early 2026)
- Zero-click search accelerating: users get answers in SERP, do not click through
- SEO impact: AI Overviews replace featured snippets; favor brand entities, structured content, authority signals; 93.8% of linked websites come from outside the first organic results page
- Source: [Searches Everywhere — Google AI Overviews Early Data 2026](https://www.searcheseverywhere.com/blog/google-ai-overviews-in-2026-search-data)
- Source: [EnFuse Solutions — Google AI Overviews Changing SEO 2026](https://www.enfuse-solutions.com/how-googles-ai-overviews-are-changing-seo-in-2026/)

#### Pricing
- Integrated into free Google Search; no separate subscription
- Google AI Pro ($19.99/month) unlocks higher AI Mode limits, Gemini 3.1 Pro in AI Mode, Nano Banana Pro images

---

### 14. You.com

**Category:** Privacy-first AI search engine / Multi-mode AI assistant  
**Competitive Position:** Niche privacy play; pivoting from consumer to enterprise B2B; $1.5B valuation

#### Current Status
- Raised **$100M Series C at $1.5 billion valuation** (September 2025; led by Cox Enterprises with Georgian, Salesforce Ventures, Norwest)
- Pivoting from general AI search to enterprise/business-focused AI assistant (shifting from B2C to B2B)
- You.com's differentiation: Privacy-first (no surveillance/targeted ads), customizable multi-mode interface, no data retention by default
- Source: [The SaaS News — You.com Raises $100M Series C](https://www.thesaasnews.com/news/you-com-raises-100m-series-c-at-a-1-5b-valuation)
- Source: [The Information — You.com Completes Cox-Led Fundraising](https://www.theinformation.com/briefings/com-completes-cox-led-fundraising-1-5-billion-valuation)

#### Key Capabilities Shipped
- **Multi-mode search interface:**
  - Smart mode: Quick AI-powered answers
  - Research mode: Deep analysis with multiple source citations
  - Genius mode: Advanced reasoning for complex problems
  - Create mode: AI content generation
- **Apps feature:** Specialized content blocks in search results (modular)
- **Privacy:** No targeted ads; optional zero data retention; no search history stored by default
- Enterprise focus: Internal knowledge search, file repository, private spaces

#### Pricing
| Plan | Price | Notes |
|------|-------|-------|
| Free | $0 | Basic search with AI |
| YouPro | $15/month (annual) | Advanced AI models; productivity features; ~$5 cheaper than GPT/Claude Pro |

- Positioning: Slightly below ChatGPT/Claude at $20/month, targeting users who want both privacy and AI capability
- Source: [stackmatix.com — You.com AI Search Review 2026](https://www.stackmatix.com/blog/you-com-ai-search-review)

---

## QUICK REFERENCE: COMPETITIVE POSITIONING MAP

| Platform | Category | Primary Moat | MAU / Key Metric | Pricing Entry |
|----------|----------|--------------|------------------|---------------|
| ChatGPT | General assistant | Scale + distribution + ACP commerce | 900M weekly | $0 free / $20 Plus |
| Claude | General assistant + enterprise agent | Best-in-class coding (SWE-bench #1); enterprise trust | Undisclosed | $0 free / $20 Pro |
| Gemini | General assistant + Google ecosystem | Google integration depth; 750M MAU; search distribution | 750M MAU | $0 free / $19.99 AI Pro |
| Grok | General assistant + real-time social | X platform distribution; real-time data; fastest US growth | 78M MAU | X Premium $8/mo |
| Perplexity | Search-native + enterprise research | Citation accuracy; multi-model orchestration; Computer | ~30M MAU / $200M ARR | $0 free / $20 Pro |
| DeepSeek | Open-weight model provider | Price-performance (5–50× cheaper than US models); open weights | ~97M MAU (peak 2025) | Free consumer / $0.28/M API |
| Meta AI | Embedded assistant + open-source | 1B MAU social distribution; Llama open-weight ecosystem | 1B+ MAU | Free |
| GitHub Copilot | Coding assistant | 42% market share; VS Code/GitHub distribution; enterprise | 20M users | $0 / $10 Pro |
| Cursor | Agentic IDE | $2B ARR growth; Automations; enterprise adoption | $2B ARR / ~$50B valuation | $20 Pro |
| Replit | Cloud IDE + agent | No-install browser IDE; Agent 4 parallel tasks; non-coder access | Enterprise case studies | Free / $20 Core |
| Windsurf | Agentic IDE | Cascade Flow autonomy; auto-context; value pricing | $2.85B valuation (2025) | $15 Pro |
| Google AI Overviews | AI search layer | 2B user search distribution; Google index | 2B monthly via Search / 75M daily AI Mode | Free (part of Google Search) |
| You.com | Privacy AI search | No-surveillance search; multi-mode; B2B pivot | Undisclosed | Free / $15 YouPro |

---

*Compiled March 22, 2026. Sources linked inline. All statistics from reported dates indicated. DeepSeek V4, Grok 5, and several announced products pending release at compilation date.*