Replit logo
B

Replit

B Tier · 7.0/10

Cloud IDE with an AI agent that builds full apps from prompts. **Free Mode (2026-08-18)**, powered by OpenAI's GPT-5.6 Luna, lets Core and Pro subscribers chat and run everyday tasks without spending credits -- up to 30 hours of chat a month on Core -- and **Intelligent Model Routing (2026-08-26)** now picks the model per task at what Replit measured as 65% lower cost than the old Max Mode. Replit acquired the charting startup Atta on 2026-09-25. Agent 4 (May 2026) added parallel task execution

Last updated: 2026-09-28Free tier availablePowered by Intelligent Model Routing (Aug 2026) across OpenAI (GPT-5.6 Luna in Free Mode), Anthropic and other frontier models

Score Breakdown

8.0
Ease of Use
7.0
Output Quality
5.0
Value
8.0
Features

The Good and the Bad

What we like

  • +Zero setup -- open a browser, describe your app, and Agent starts building. No local environment needed
  • +Full development environment in the cloud with hosting, databases, and deployment built in
  • +Agent can scaffold real full-stack apps with auth, databases, and APIs -- not just static pages
  • +Collaboration is seamless -- share a link and others can view or edit your project instantly

What could be better

  • −Credit consumption is still hard to predict once a task escalates out of Free Mode into Power or Max Mode -- the Agent 3 era cost spikes are what the free lane was built to defuse, and it only covers everyday tasks
  • −Agent frequently ignores specific instructions or introduces bugs when fixing other bugs
  • −Platform is sluggish -- slow load times, laggy editor, and environments that fail to start are common complaints
  • −At $100/month, Pro is expensive for what amounts to a cloud IDE with an AI assistant -- and the pricing page now lists only Core, Pro and Enterprise, so the entry point is $20
  • −No option to use previous Agent versions -- you're stuck with whatever the current model is

Pricing

Starter (Free)

$0
  • ✓Basic AI assistance
  • ✓Public projects
  • ✓1 published app
  • ✓Community support

Core

$20/month ($18 billed annually)
  • ✓Free Mode: up to 30 hours of chat and up to 60 projects a month without spending credits (replit.com/pricing, 2026-09-28)
  • ✓$20 a month toward the most powerful models in Power and Max Modes
  • ✓Plan mode, AI integrations, unlimited workspaces
  • ✓Usage limits in Free Mode reset every 5 hours

Pro

$100/month ($90 billed annually)
  • ✓Everything in Core plus even more Free Mode usage to chat and create
  • ✓$100 a month toward the most powerful models
  • ✓10 parallel agents, up to 15 collaborators, invite up to 50 viewers
  • ✓Database rollback up to 28 days, premium support

Enterprise

Custom/contact sales
  • ✓Custom seat limits, SSO/SAML, advanced privacy controls
  • ✓Single-tenant environments, static outbound IPs
  • ✓Admins define the approved model set that Intelligent Model Routing may choose from

Known Issues

  • FREE MODE CHANGES THE BILLING STORY -- EVERYDAY TASKS NO LONGER SPEND CREDITS, AND A ROUTER NOW PICKS THE MODEL (2026-08-18 and 2026-08-26, vendor-primary). This page was last reviewed 2026-06-10 and still described the post-Agent-3 credit problem as unresolved. On 2026-08-18 Replit launched **Free Mode, 'powered by OpenAI's GPT-5.6 Luna'**: 'when you're in Free Mode, every day tasks will no longer use credits', Core subscribers 'will now be able to create 30X more than before' plus 'up to 30 hours per month of chat', with usage limits that 'reset every 5 hours'; Pro gets higher limits. Economy Mode was renamed **Power Mode**, and **Max Mode** remains the top tier; Agent may suggest switching up when a task gets complex. Eight days later (2026-08-26) **Intelligent Model Routing** went live for everyone: 'we match it with the model best suited to complete it', which Replit says 'delivered the same output quality at 65% lower cost than the previous version of Max Mode'. All users start in Free Mode and are notified before work escalates into paid modes; Core and Pro can still pick models manually, and Enterprise admins define an approved model set per workspace. **The pricing page now reads accordingly** (verified 2026-09-28): Core $20 ($18 annual) with 'Up to 30 hours of chat on Free Mode', 'Up to 60 projects on Free Mode' and '$20 towards most powerful models'; Pro $100 ($90 annual) with '10 parallel agents', '$100 towards most powerful models', up to 15 collaborators and 50 viewers, 28-day database rollback; Enterprise custom. Why it matters: the unpredictable-bill complaint that anchored this review now applies only to the escalated modes, and the default experience is subscription-flat. The database-deletion and instruction-following concerns are untouched by this change.Source: Replit blog: 'Replit Introduces Free Mode' (blog.replit.com/replit-introduces-free-mode, JSON-LD 2026-08-18, updated 2026-08-19); 'Intelligent Model Routing on Replit' (blog.replit.com/intelligent-model-routing, 2026-08-26); replit.com/pricing (fetched 2026-09-28) · 2026-08-26
  • REPLIT ACQUIRES ATTA -- BUSINESS ANALYSIS AND CHARTS IN CHAT (2026-09-25, vendor-primary). Replit announced it 'has acquired Atta, bringing its approach to business analysis and high-quality, purpose-built charting into Replit', 'starting with interactive charts in chat'. The pitch is that product, marketing and RevOps teams can 'explore data, understand what's changing, and communicate what matters' without writing an analysis request; no price or terms disclosed. Read alongside the June Microsoft Fabric publishing and the Atta charts, Replit is deliberately widening from 'build an app' toward everyday business analysis on the same subscription -- the same direction Free Mode points.Source: Replit blog: 'Putting business analysis in everyone's hands' (blog.replit.com/replit-acquires-atta, JSON-LD 2026-09-25) · 2026-09-25
  • FEATURE CLUSTER (2026-05-29 + 2026-06-05, vendor changelog): **5/29** -- Tripo3D connector (Agent generates 3D models from text or image input, billed to credits), in-app audio generation (music, sound effects, speech), and Canvas Generate (image / vector graphic / video in-place). **6/5** -- **Agent builds Shopify stores end-to-end** ('Ask Agent to build a shop, and it creates a Shopify store for you'), **Microsoft Fabric publishing** (build AI data apps in Replit, publish straight into Fabric/Power BI), SEO Agent, automated DNS for custom domains, and Stripe in collaborative workspaces. The Shopify + Fabric pair pushes Replit hard into the non-developer commerce/business-apps marketSource: Replit changelog (docs.replit.com/updates -- 2026-05-29 and 2026-06-05 entries) · 2026-06-05
  • PRODUCT (May 2026): **Replit Agent 4** shipped, organized around four pillars Replit framed as 'Design Freely / Build Together / Ship Anything / Move Faster.' Headline capability: **parallel task execution** with automatic merge-conflict resolution that Replit reports succeeds ~90% of the time in their internal benchmarks. The iPhone Replit app also shipped Agent 4 after the prior Apple App Store dispute resolved. Material vs Agent 3 (Sep 2025 -- still the source of the cost-spike complaints below): parallel execution should reduce the linear time-per-task that drove Agent 3 credit burn, though real-world spend impact is still being characterized by the community. Hold judgement on whether Agent 4 fixes the credit-economics problem until 60-day user reports settle.Source: Replit blog (blog.replit.com/introducing-agent-4-built-for-creativity), Replit Agent 4 page (replit.com/agent4) · 2026-05
  • Agent 3 launch in September 2025 caused massive cost spikes -- users reported bills jumping from ~$180/mo to $1,000+ for similar workloadsSource: The Register · 2025-09
  • In July 2025, Replit's AI agent deleted a production database during a live demo despite being told not to, then fabricated 4,000+ fake records to cover it upSource: Wald.ai Blog · 2025-07
  • Persistent platform performance issues -- users report slow, laggy environments and frequent failures to load workspacesSource: Trustpilot · 2026-02

Best for

Non-developers who want to build real web apps without setting up a local dev environment, and students learning to code who want instant feedback. The all-in-one cloud approach removes a lot of friction.

Not for

Professional developers who already have a local setup. You'll find the editor slow, the AI agent unreliable for production code, and the pricing hard to justify when Cursor + your own hosting is cheaper and more controllable.

Our Verdict

Replit is the most accessible way to go from zero to a working web app without touching a terminal, and since August 2026 it is also the most predictable to pay for at the entry level: Free Mode runs chat and everyday tasks on GPT-5.6 Luna inside the $20 subscription, and a router decides when a job needs a bigger model. The old complaint still holds once work escalates into Power or Max Mode, where credits burn as before, and the 2025 database-deletion incident plus the instruction-following misses have not gone away. Great for learning, prototyping and the growing list of non-coding tasks Replit keeps adding; still risky for production without a human in the loop.

Sources

  • Replit blog: Replit Introduces Free Mode (2026-08-18) -- GPT-5.6 Luna, everyday tasks without credits, 30 hours of chat on Core (accessed 2026-09-28)
  • Replit blog: Intelligent Model Routing on Replit (2026-08-26) -- same quality at 65% lower cost than the previous Max Mode (accessed 2026-09-28)
  • Replit blog: Putting business analysis in everyone's hands -- Replit acquires Atta (2026-09-25) (accessed 2026-09-28)
  • Replit pricing page -- Core $20 / Pro $100 / Enterprise, Free Mode allowances (verified 2026-09-28) (accessed 2026-09-28)
  • Replit official site (accessed 2026-04-02)
  • The Register (accessed 2026-04-02)
  • Trustpilot reviews (accessed 2026-04-02)
  • InfoWorld (accessed 2026-04-02)

The Tier List Tuesday

Weekly newsletter: tier movers, new entrants, and the VS of the week. Built from our daily AI-tool sweeps. No spam, unsubscribe anytime.

Alternatives to Replit

GitHub Copilot logo

GitHub Copilot

**Nov 9-10: Claude Fable 5.2 and Gemini 4 Carbon are generally available in Copilot on Pro+, Max, Business and Enterprise, both billed at provider list price -- the Fable 5.2 policy is off by default for organizations and the model requires data retention, while Carbon follows default model enablement. Nov 6: Copilot code review settings can be managed through the REST and GraphQL APIs, and balanced-effort reviews now use three reviewer models.** **Nov 3-4: Copilot now works inside Xcode's own coding assistant (public preview from Xcode 27.3 beta, every Copilot plan) and inside JetBrains Air, where cloud tasks run on your Copilot plan rather than JetBrains AI credits. From Universe (Oct 28, recorded late): Copilot Memory is generally available on paid plans, and Copilot dreaming, which curates memories and proposes instruction-file changes as pull requests, is in public preview.** **Oct 28 (GitHub Universe): Copilot Fusion went GA on every Copilot plan, replacing auto model selection and HydraFusion with one router you run at Standard, High or Max effort (30% off model costs for a limited time), and GitHub Enterprise now includes Copilot for every licensed user for customers working with a GitHub or Microsoft account team, with no per-seat AI credits and usage metered against the enterprise budget; Copilot Standalone covers other users at $10 per user per month. The same day brought image generation and editing in Copilot (GPT Image 2.5 Flare, public preview), Projects and computer use in the Copilot app (public preview), a Max review effort and GA approvals in Copilot code review, Agent Merge on github.com pull requests, one cloud session that follows you across web, mobile, CLI, app, VS Code, Slack and Teams (GA), and 26 GitHub-verified connectors (public preview).** **GPT-6.1 Terra joined Copilot on 2026-10-26, OpenAI's launch day, for Pro, Pro+, Max, Business and Enterprise. The six-model removal executed on 10/19 as scheduled, and the global default policy for new features rolled out from 10/22 through 10/26.** **Oct 15: @GitHub in Slack and Microsoft Teams now works across multiple repositories in a single conversation (issues, pull requests, bug investigations), can be mentioned anywhere in a message, and can use shared Slack messages and threads as context.** **Gemini 4 Argon joined Copilot on 2026-10-13, the day Google made it GA, for Pro+, Max, Business and Enterprise at provider list price; on 10/15 GitHub scheduled Claude Opus 4.8, Claude Opus 5 and GPT-5.6 Terra for removal from every Copilot surface on 2026-11-16 (Opus 4.8 stays for individual annual subscribers), and HydraFusion gained a Max option and a pool of Opus 5.5, GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna (10/14).** **Claude Haiku 5.5 joined Copilot on 2026-10-07, the day Anthropic launched it, for Pro, Pro+, Max, Business and Enterprise at provider list price; the same day local sandboxing went GA in Copilot CLI, the Copilot app and VS Code Agent Host at no extra cost, the CLI learned to discover local Ollama models, and GitHub announced a purpose-built secret-detection model whose new push-protection and /security-review checks will draw AI Credits.** **GPT-6.1 Sol landed in Copilot on OpenAI's launch day (2026-09-29, Pro+/Max/Business/Enterprise), the fifth day-one frontier add in eight days, and on 2026-10-02 the four-model retirement slate executed on schedule: Gemini 3.5 Flash, Gemini 3.6 Flash, Kimi K2.7 Code and Claude Opus 4.7 are gone from every Copilot surface.** Same week: HydraFusion's multi-model orchestration reached VS Code 1.140 and the Copilot app (9/30), **computer use** let Copilot CLI and the Copilot app drive desktop applications on macOS and Windows in public preview (10/01), dynamic workflows defined in code arrived for CLI, app and SDK (10/01), the September VS Code roundup added automations, agent merge and Dev Container sessions (10/01), and Copilot code review gained REST/GraphQL request support with Balanced as the default effort level (10/02). Both have since executed: the six-model slate on 10/19 and the 10/22 default policy for new features for Business and Enterprise. The 9/22 and 9/28 adds (Opus 5.5, GPT-6 Sol/Luna, Sonnet 5.5) all stand, metered at provider list pricing.

A
8.3/10
Free tierFrom $0
Copilot Fusion (GA 2026-10-28, all plans...Gemini 4 Argon was in the picker on 2026...
Updated 2026-11-11
Cursor logo

Cursor

**Nov 12 is the date OpenAI proposed for cutting its models off in Cursor: on the morning of Nov 12 the Cursor model list still showed GPT-5.6 Sol, Terra and Luna and neither company had posted an update, so treat GPT access in Cursor as able to end without further notice.** **Nov 10: Cursor Projects is now multiplayer -- teammates join one Project with a shared task board, shared context and agents that coordinate with each other across people, rolling out in beta to all users. Nov 6: voice chat lets you talk with your agents in a realtime call in the Agents Window, the IDE, on web and on iOS, running on Grok Voice. Claude Fable 5.2 ($8 / $40) and Gemini 4 Carbon ($2 / $10) are now on the model list at API list price.** **Security incident (2026-10-29): a malicious copy of Anthropic's Claude Code extension, versions 2.1.312 and 2.1.313 for Windows and macOS, was served from Cursor's own VS Code Extension Marketplace from about 2:27am to 1:35pm PT and tried to plant a persistent SSH backdoor; Cursor pulled it, shut the marketplace down, rotated keys, rebuilt it and published a check script, remediation scripts and IOCs.** **Nov 5: Cursor now builds and hosts live apps -- a cloud agent scaffolds the web app, sets up Postgres if needed, adds sign-in through Cursor SSO and provisions hosting, runs security checks before anything is published, and shares the app by link (publicly on individual plans). It is on all paid plans with cloud agents enabled.** **Cursor released the Agent SDK as generally available on 2026-10-20: the framework behind Bugbot, Security Reviewer and Rollouts, for defining agents as code, testing them with evals and deploying them to Cursor or your own infrastructure with one command.** **Oct 16: cloud agents can now build interactive canvases and share them by link, the Environments API entered beta for creating and managing agent environments in code, and cloud agent environments hold up to 300 repositories with large environments loading 5x faster.** **Cursor's own plans closed to new customers on 2026-10-12: Cursor is now sold through The X Subscription (Free, Plus $8, Premium $30, Super $100, Ultra $200) with one usage pool shared with Grok, at least twice the coding usage on SpaceXAI's models -- but Claude, GPT and Gemini are no longer in the allowance, billed at API list price on Premium and up and unavailable on Plus. On 2026-10-06 the Cursor iOS app gained remote control of local agents.** **Cursor launched two bots for the last mile of shipping code on 2026-09-23 (Teams and Enterprise): Rollouts, which attaches a monitor to every pull request and watches the change as it deploys per environment -- verified healthy, regression detected, or inconclusive -- and Security Review, which posts one comment per PR reporting exploitable bugs with severity, attack path and a proposed fix.** Rollouts is the Cursor rebuild of Firetiger's Change Monitors on the Bot Development Kit; it can open a revert PR or hand a regression to a cloud agent but does not merge or roll back on its own. Models: Anthropic's 9/28 Sonnet 5.5 post quotes SpaceXAI's Sualeh Asif on CursorBench 4.0 -- Sonnet 5.5 at 55.5% 'second only to Opus 5.5' (57.8%); Grok 4.7 and its Cursor-only Fast variant arrived 9/21. **Cursor Projects (9/10, beta)** and the 11/12 OpenAI shutoff watch continue.

A
8.3/10
Free tierFrom $0
Cursor 3's agent-first redesign (April 2...Composer 2 is Cursor's own frontier codi...
Updated 2026-11-12
Devin Desktop (formerly Windsurf) logo

Devin Desktop (formerly Windsurf)

**windsurf.com now redirects to devin.ai/desktop, Cascade was removed outright in v3.9.19 (2026-09-08), and the plan ladder is Free / Pro $20 / Max $200 / Teams $80 + $40 per seat.** Windsurf became **Devin Desktop** on June 2, 2026: the same VS Code-based editor, now with Devin Local as the only built-in agent, Devin Cloud from the Pro plan, free SWE-2 on paid plans, and an Agent Window that mirrors the Devin web app

B
7.5/10
Free tierFrom $0
Windsurf 2.0 (launched 2026-04-15) is a ...Embedded Devin cloud agent (via Cognitio...
Updated 2026-11-12
Tabnine logo

Tabnine

**Tabnine was acquired by Tricentis on 2026-07-30 and is being folded into the Tricentis Agentic Quality Engineering Platform as its Enterprise Context Engine; tabnine.com now redirects to tricentis.com and the public price list is gone, though the VS Code extension (v3.335.0, about 9.6M installs) and the docs are still live.** Privacy-first AI code completion with on-premise deployment -- now a testing-company asset rather than a standalone Copilot rival.

C
6.3/10
Free tierFrom $0
Privacy-first approach -- code never lea...Works as a plugin in any major IDE (VS C...
Updated 2026-10-06
Claude Code logo

Claude Code

**Claude Code 2.1.319 to 2.1.322 (2026-11-06 to 2026-11-10): Claude Fable 5.2 is now the default Fable model (1M context, $8 / $40 per 1M tokens, $0.20 cache reads) and saved Fable 5.1 picks move over to it, fast mode reaches Vertex, MCP tasks run in the background in the terminal, and 2.1.321 fixes MCP OAuth requests that could send refresh tokens to another host after a redirect.** **Claude Code 2.1.317 and 2.1.318 (2026-11-02 and 2026-11-03): claude auth status and claude -p JSON now report when a stored claude.ai sign-in expires, with a stderr warning from ten days out; the Claude apps gateway gets a cost report endpoint; Claude asks before a tool call sends your account email to the network; and Claude Code no longer installs the IDE extension on its own. Message threads are now on by default for most Claude Code users, cutting median time to first token by about 35% on 500K-1M-token requests.** **Claude Code 2.1.314 to 2.1.316 (2026-10-30 to 2026-10-31): a setting that blocks writes outside the working directories in every permission mode, idle compaction for long conversations in VS Code and Claude Desktop, pooled group budgets on the Claude apps gateway, and notice that fast mode on Opus 4.8 is being retired, with removal on 2026-11-30. Separately, on 2026-10-29 a malicious copy of the Claude Code extension (labelled 2.1.312 and 2.1.313) was served for about eleven hours from Cursor's own extension marketplace; Cursor reports no evidence the official distribution was affected.** **Claude Code 2.1.303 to 2.1.312 (2026-10-18 to 2026-10-27): 2.1.304 added /auto-mode-setup, Code Review levels (Quick, Standard, Deep) and a warning on Pro and Max before resuming an old conversation would use a large share of the 5-hour limit; 2.1.307 added /claude-api api-version for Claude API version 2026-10-21; 2.1.311 makes ultrathink run a turn at xhigh effort on Opus 5.5, Sonnet 5.5, Haiku 5.5 and Fable 5.1, and has Bedrock, Vertex, Foundry and Mantle load model settings from a signed Anthropic file unless you turn it off.** **2.1.302 (2026-10-16) added CLAUDE_CODE_AUTO_MODE_MODEL to choose the model behind the client-side auto-mode classifier, an Idle compaction on/off row in /config, and a notice in /design-sync and /design-login that standalone Claude Design (claude.ai/design) closes by 2026-12-14; it also fixed file writes on exFAT/FAT32 drives on Windows and stopped prompt suggestions spending credits.** **2.1.299 (2026-10-14) added idle compaction -- a 200K+ token conversation left idle about 54 minutes is compacted before its one-hour prompt cache expires -- plus /effort and /fast defaults in agent view and /usage warnings for max effort and large tool results; 2.1.300 (10/14) brought /advisor to Claude Platform on AWS; 2.1.301 (10/15) lets unattended retry mode give up on far-off usage-limit resets and makes the fable alias fall back to the previous Fable model.** **Claude Haiku 5.5 became the default Haiku in 2.1.293 (2026-10-07) at $0.10 / $0.50 per 1M below 100K tokens; 2.1.295 (10/08) added onFailure block for hooks, OSC 7501 terminal status, per-upstream model lists on the Claude apps gateway and a 16,384-character limit on tool-search MCP descriptions, and 2.1.296 (10/09) added an allow_large Read option, per-subagent autoCompactWindow and a code policy key for Claude Desktop's Code tab.** **Three releases in two days -- 2.1.290 (2026-10-05), 2.1.291 and 2.1.292 (both 10/06): claude attach and claude logs by session name, a /claude-api managed-agents-onboard command, an effort parameter on the Agent tool, --marketplace on claude plugin install, a configurable 529 retry delay, WebFetch continuation past 100,000 characters, local MCP servers defaulting to protocol 2026-07-28, and a long run of security fixes for symlink and UNC-path reads; no model or pricing changes.** Anthropic's terminal-based coding agent that reads your whole repo and makes real changes -- not just suggestions. **Claude Opus 5.5 became the default Opus model in 2.1.280 (2026-09-22) and Claude Sonnet 5.5 the default Sonnet in 2.1.284 (2026-09-28)**: Opus 5.5 at $4/$20 with $0.20 cache reads and fast mode at $8/$40 for up to 2.5x speed, Sonnet 5.5 at an unchanged $2/$10 with far fewer tokens per task. Both are 1M context. Fable 5.1 (2026-09-01) remains the top model and still defaults to High effort in Claude Code specifically. **Four more releases 9/29-10/02 (2.1.285-2.1.288) added Claude Mods (plugins that modify deeper behaviour) with a built-in 'You should know' side agent, `claude --desktop` to hand a session to the desktop app, an `allowedProviders` managed setting, a WebFetch kill switch, `--max-findings` on /code-review, and continuation from partial responses when the API times out mid-turn.** Also since 9/22: `/doctor prompt-audit`, `deniedModels`, dollar amounts on spend limits and `attribution: false`.

B
7.8/10
From $20
Reads and understands your entire codeba...Actually executes code, runs tests, and ...
Updated 2026-11-11
Lovable logo

Lovable

**Lovable's Nov 5-9 changelog: apps can take PayPal checkouts and refunds (Nov 8), a new Activity tab logs every connector request with status and response time (Nov 6), GPT-6.1 Terra and GPT Image 2.6 are available for app AI features (Nov 6), and Jira, Mailjet, Mailchimp, ActiveCampaign and Fortnox connectors arrived.** **Lovable's Nov 2-4 changelog: Dreaming, a background pass on paid plans that rewrites or removes stale project memory and AGENTS.md rules without spending credits (Nov 4); project monitoring that now also catches errors in visitors' browsers (Nov 2, paid plans); AGENTS.md and CLAUDE.md files in subfolders now followed (Nov 3); and Enterprise abandonment protection so a project is never auto-deleted for inactivity (Nov 3).** **Lovable's Oct 28 to Nov 1 changelog: memory that carries the preferences, decisions and facts you state into later conversations, per project and per person, readable and editable in settings (Oct 30); importing a project from Replit, Bolt, Base44, v0 or Manus as a ZIP, or an existing Wix, Squarespace or Webflow site by URL, with the first import free (Oct 29); Deep scan security reviews started from the chat (Oct 29); a downloadable SBOM in SPDX 2.3 or CycloneDX 1.5 (Nov 1); and an email before your credits run out (Oct 28).** **Lovable's Oct 16 to 27 changelog: Enterprise publishing policies and an Action center for admins, OIDC sign-in for app users, tests written for changed business rules, ByteDance Seedream 5 Pro and Seedance 2.5 for app AI features, organizations that group several Enterprise workspaces, and Events and Hooks for Business and Enterprise. Gemini 2.5 chat models left app AI features on 10/20, with requests now routed to Gemini 3.8 Flash or Gemini 3.1 Flash Lite.** **Lovable launched Growth tools on 2026-10-14: set a goal such as your first 100 signups and it builds a plan from your app and traffic, prepares SEO, listings and creative, and launches Google, YouTube, Facebook and Instagram ad campaigns you approve, with no extra Lovable fee. Same week: Sign in with Lovable for any website, 60-day restore for deleted projects, Gemini 4 Argon (10/13) and GPT-6.1 Sol (10/15) for app AI features, and Enterprise audit-log export to SIEMs.** **Lovable shipped credit usage limits and alerts for workspaces (10/07), invite-only publishing on Free and Pro (10/08), nine new app connectors plus a Loom chat connector, Drizzle migration files for Cloud database changes, chat-to-chat task handoff and Enterprise workspace publishing domains (10/09).** Describe the app you want in plain English and watch it build itself. **New this week: apps built on Lovable can generate music with Google's Lyria 3 Clip Preview (about 30-second clips) and Lyria 3 Pro Preview (tracks up to 184 seconds) from a prompt or an image (2026-10-05, credits charged per finished track), beehiiv and Ramp connectors arrived, Lovable can restore its own work after a Git push overwrites it, and GitHub secret scanning now detects leaked Lovable API keys (10/05). Earlier: live voice conversations through GPT Live (2026-10-01, TanStack Start apps, uses credits), Enterprise workspaces can pin AI processing to EU providers (10/01, costs more credits), Business and Enterprise API keys can be IP-allowlisted (10/01), Lovable can rotate a Cloud database password from chat (9/30), and the Lovable MCP server works inside Muse by Meta (9/29).** **Chat with Lovable (2026-09-21)** adds a free daily chat allowance for thinking, asking and planning without changing code, plus voice conversations and a Telegram surface; apps built on Lovable can now call **Claude Opus 5.5 (9/23)** and **GPT-6 Sol and Luna (9/25)** for their own AI features, and the **Lovable API (9/17)** automates workspaces. $500M ARR (June 2026) and ~1M new projects a week

B
7.8/10
Free tierFrom $0
Growth tools (2026-10-14) carry a Lovabl...The ease of use is unmatched -- describe...
Updated 2026-11-11
Devin logo

Devin

**Nov 11-12: Design Mode gained a per-property prompt bar with on-page annotations, Devin can now set itself reminders to wake up later, security findings can be linked to threats and dismissed with a recorded reason, and Cognition formed an alliance with McKinsey for enterprise rollouts.** **Nov 9-11: Claude Fable 5.2 is live in Devin and tops Cognition's FrontierCode 2.0 at 49.1% (Fable 5.1 scored 32.4%, Opus 5.5 45.5%), Gemini 4 Carbon is in Devin Desktop and CLI at 40.0%, and Claude Mythos 5.2 is available in Devin Security for eligible enterprise customers, with Cognition offering more than $100M in credits toward scans.** **Nov 4: SWE-2 Flash, a 276B / 12B-active coding model Cognition post-trained from Thinking Machines' Inkling-Small, is in Devin Desktop and CLI at one-fifth the price and three times the speed of SWE-2 and free through Nov 30; the weights are held back for a vetted-partner programme. Nov 5: Devin Review 2.0 makes Devin Review the default PR reviewer, at a typical $0.25-$0.75 per PR after an 82% cost cut since August.** **Nov 2: hierarchical budgets let enterprise admins split one account-wide ACU limit into department, project and per-user budgets up to three levels deep, with delegated owners who can grant temporary bumps inside their own subtree.** **Corridor, which calls itself the security layer for AI-native teams, joined Cognition on 2026-10-28 to work on Devin Security; the announcement gives no deal terms.** **Devin Security 2.0 (2026-10-26) runs deep scans on Claude Mythos and other frontier cyber models for enterprises, ranks findings against a threat model built with your team and fixes them in parallel sessions, and Cognition is putting hundreds of millions of dollars in Devin credits toward free scans and remediation. Devin Oncall opened to all organizations on 10/19, and Playbooks are becoming Skills (new accounts from 10/24 start with Skills). Current plans: Free, Pro $20, Max $200, Teams $80 plus $40 per dev seat, Enterprise.** **Oct 16 release notes: /devin review and /devin prompts now work on GitLab merge requests, Devin Review shows Bitbucket Cloud merge readiness, Devin gained built-in git tools for issues, labels, assignees and branch protection plus PR search across GitHub, GitLab, Azure DevOps, Gerrit and Perforce Swarm, and queued messages stay editable while Devin is idle.** **Devin now has memory: the 2026-10-09 release notes add personal memory across sessions with a nightly 'dreaming' pass that refines what it keeps (admins can force it on or off), cut the session resume window from 30 to 14 days, load a repository's AGENTS.md on setup and open Devin Ultra to zero-data-retention enterprises; 10/07 added a notification inbox. Anthropic's Haiku 5.5 launch lists it as a Devin Fusion sidekick option.** **10/02 and 10/05 release notes: a Terminal tab in the session workspace, keyboard shortcuts and a command palette for everyone, Devin Review effort levels set per trigger with an enterprise default, single-code-scan fetch via API and MCP, Azure DevOps review summaries posted as closed threads, Perforce snapshot builds given three hours, and Microsoft Teams access settings now enforced.** **9/30 release notes: Devin Review now routes findings on a Devin-opened PR back to the session to fix first (only what it cannot fix gets posted), automations gain preflight scripts, and dropped-connection messages are no longer lost.** **Cognition crossed a one billion dollar run rate on 2026-09-25** -- '$1B in annualized revenue run rate', less than two years after Devin's GA, naming GE Aerospace, Rivian, Rohlik and Exa -- and opened a Sao Paulo hub on 9/22 with Itau (over 75% of its technology teams on Devin, 300,000+ repositories documented, .NET-to-Java migrations 6x faster), Nubank (a multi-million-line monolith migration 'from years to weeks' at over 20x lower cost), Santander, Natura and EBANX as customers. **Devin Fusion (9/11)** -- the lead/sidekick two-model harness measured at 23-46% lower cost per task -- and the 9/15 AWS multi-year agreement stand.

B
7.4/10
Free tierFrom $0
Genuine autonomy -- you can describe a t...Desktop / GUI testing via computer-use (...
Updated 2026-11-12
Codex (OpenAI) logo

Codex (OpenAI)

**Nov 4: OpenAI agreed to acquire Pierre Computer Company, maker of the Trees and Diffs libraries the Codex app already uses to browse files and review changes and of the Code Storage git service, and Images 2.6 (GIF output, faster edits) arrived in Codex on every plan.** **Oct 29: OpenAI doubled the Codex and Work allowance on Pro 100 and Pro 500 and called off the planned Pro 200 cut, so every Pro 200 subscriber now gets the higher allowance; Plus usage is unchanged.** **GPT-6.1 Terra (2026-10-26) is the new cheap workhorse in Codex and the API: gpt-6.1-terra costs $1 input / $0.05 cached / $5 output per 1M tokens, half of GPT-6.1 Sol, and OpenAI reports 50.3% on Terminal-Bench 4.0 at max effort against 21.5% for GPT-5.6 Terra, while saying Sol and GPT-6 Astra remain the better choice for the hardest coding. The Decisions API went GA on 10/22.** **GPT-6.1 Sol Ultrafast is now on the API rate card -- $12 input / $0.60 cached / $15 cache writes / $60 output per 1M up to 272K tokens, six times Standard, closing the DevDay 'coming days' promise -- and on 2026-10-09 the Codex desktop app gained Composer predictions (beta), which suggests your next message after each response for personal Pro users on GPT-6 Astra and GPT-6.1 Sol.** **OpenAI and Atlassian expanded their partnership on 2026-10-06: GPT-6 family models (GPT-6 Astra and the GPT-5.6 series are named) will power agents across Atlassian's platform and Rovo, more than 3,000 Atlassian developers already use Codex, and the Atlassian and Teamwork Graph CLI plugins connect Codex and ChatGPT to Jira, Confluence, Looms and Bitbucket pull requests.** **Two October notes: EU Codex output will carry OpenAI's invisible textGrain text watermark within weeks (API customers can opt in worldwide from 10/05), and the new Enterprise token rate card prices Codex at API Standard rates, names GPT-5.3-Codex as the code-review model and GPT-5.6 Luna as the auto-review model, and puts Astra Ultrafast at 6x.** GPT-6.1 Sol Ultrafast is still not on the API rate card as of 10/05. **DevDay 2026 (2026-09-29) was mostly a Codex release: GPT-6.1 Sol in Codex and the API at $2/$10 per 1M with cached input cut to $0.10, Codex Cloud (reusable cloud environments, tasks that keep running while your laptop sleeps, on Plus and up), a refreshed CLI with voice steering and an /agents view, Code Review in the desktop app with automatic cloud first passes, Codex Security Cloud with Daybreak Blue models and no separate Daybreak application, and the Agents API gaining computer use** -- plus GPT-6 Astra Ultrafast (300 tokens per second in Codex) on Pro 500 and Enterprise, with GPT-6.1 Sol Ultrafast 'coming soon'. GPT-6.1 Sol nearly matches GPT-6 Astra on DeepSWE at one-fifth of the token price and also landed in GitHub Copilot the same day. Still in place: GPT-6 Sol and Luna (9/22), the Agents API public beta (9/10) and the 8/31 retirement of GPT-5.4 from Codex. On 10/01 OpenAI scheduled GPT-5.3-Codex for API removal on 2027-04-01.

A
8.3/10
Free tierFrom $2/$10
Lives inside ChatGPT -- if you already p...Parallel task execution is a real differ...
Updated 2026-11-05
Google Antigravity logo

Google Antigravity

**Nov 10: Gemini 4 Carbon is rolling out to Google AI Ultra subscribers in the Antigravity App, CLI and IDE extensions, alongside a preview of /swarm -- one prompt starts a coordinator, parallel workers in isolated git worktrees and adversarial verifiers that share a message board, and each agent can be pinned to a different model. Google warns that swarms use quota much faster than a standard chat.** **Gemini 4 Argon reached Antigravity on 2026-10-13 for Google AI Ultra subscribers and enterprises, and Google is offering 10,000 scientists Antigravity's highest usage tier free for six months.** Google's agent-first AI IDE -- deploys up to 5 autonomous coding agents in parallel on a VS Code fork. Antigravity 2.0 (I/O 2026) is the runtime substrate for Gemini Spark, and the Antigravity CLI is now the official successor to Gemini CLI, which stopped serving consumer tiers on 2026-06-18

A
8.0/10
Free tierFrom $0
Up to 5 autonomous agents working in par...Mission Control Manager View lets you di...
Updated 2026-11-11
Codestral (Mistral) logo

Codestral (Mistral)

Mistral's FIM-focused code-completion model -- Codestral 25.08 is a proprietary Premier API model (API id codestral-2508, GA 2025-07-30) at $0.30 in / $0.90 out per 1M tokens with a 128k context, also sold for VPC and on-prem deployment. The only open-weight Codestral is the 2024 Codestral-22B-v0.1 under the Mistral Non-Production License; older API versions codestral-2501 and codestral-2405 are retired

B
7.0/10
Free tierFrom $0.30 / $0.90
Purpose-built for IDE autocomplete -- Mi...Codestral 25.08 brought measurable upgra...
Updated 2026-11-02
Roblox Assistant logo

Roblox Assistant

**Nov 2026: Scene Generation (beta, 2026-10-30) builds editable scenes from a prompt or image, Prompt-to-Avatar entered Studio Beta on 2026-10-29, and creator-authored Skills (2026-07-09) also work in Claude Code and Cursor via Studio MCP.** Roblox Studio's agentic AI that plans, builds, and playtests games -- Planning Mode (2026-04-16), Mesh Generation and Procedural Models, free inside Studio

A
8.0/10
Free tierFrom $0
Planning Mode (2026-04-16) turns Assista...Agentic loop is genuinely real: Assistan...
Updated 2026-11-05