AI coding assistants, compared
Nine AI coding assistants compared by free plan, the lowest-priced paid plan for one person, how usage is priced and what each plan includes, where each one runs, which models it uses and whether you can bring your own key, agent features, team and admin plans, what each provider states about your code and data, and support. Every price says the plan, billing term and usage it covers, and the date it was read. Provider statements are the providers' own words, attributed to them.
What do you need?
Options
Amazon Q Developer
Online
Amazon Q Developer lists a Free Tier with 50 agentic requests per month and a Pro Tier at $19 per month per user. Its pages name plugins for JetBrains IDEs, VS Code and Visual Studio and a command line tool. AWS's product page carries an end-of-support notice for the IDE plugins dated April 30, 2027.
Our read Relevant if you want a free tier, a single per-user paid price and editor plugins for VS Code, JetBrains or Visual Studio.
- A free plan with no end date
- Runs in the editor you already use
- Free plan
- A Free Tier is listed: "Advanced capabilities at zero cost", with 50 agentic requests per month (Q&A chat and agentic coding in the IDE plugins and the CLI) and Java upgrade transformation of up to 1,000 lines of code per month. The page says the Free Tier is perpetual with monthly limits, for AWS IAM users or AWS Builder ID users, and that Free Tier limits in the IDE are for Builder ID users only.The page is undated. The Free Tier summary also lists "Access latest Claude models" and use in the IDE or CLI.Verified · Nov 10, 2026
- Lowest-priced paid plan for one person
- Pro Tier: $19 per user per month (USD, US site, 1 user), billed monthly as a subscription that starts when the user first uses a Pro feature. No annual price is shown. It lists increased agentic request limits ("Included (with limits)"), Java transformation of 4,000 lines of code per user per month pooled at the AWS payer-account level (extra lines $0.003 per line), IAM Identity Center support with admin dashboards and controls, and IP indemnity.The page FAQ says the monthly fee is charged per user and shown daily as a fraction of a month, for example $19 x 1/30 of a month = about $0.63 a day.Verified · Nov 10, 2026
- How usage is priced
- Per user per month ($19) for Pro, charged only for each user who activates it; the Free Tier has no fee. Usage is counted in agentic requests (any Q&A chat or agentic coding interaction through the IDE or the CLI) and, for Java upgrades, in lines of code submitted ($0.003 per extra line on Pro).Billing detail is from the page FAQ on activating and cancelling a Pro subscription.Verified · Nov 10, 2026
What isn't clear from public sources
The pricing page shows one monthly price per user and no annual price, gives Pro request limits only as "Included (with limits)" with no number, and lists no separate team plan: the same $19 Pro Tier is billed per user and is tied to organizations through IAM Identity Center. The product page says AWS will discontinue support for the IDE plugins on April 30, 2027 and points to Kiro. No support channel is stated.
Cursor
Online
Cursor lists a Free plan with limited usage, Plus at $8 per month (Grok and Cursor models only), Premium at $30, Super at $100 and Ultra at $200 per month, and Teams at $30+ per user per month. Its docs say Cursor is sold as part of The X Subscription and show annual prices of $80 and $300 per year for Plus and Premium.
Our read Relevant if you want Cursor's own editor, a free tier, annual billing and the option to use your own API key on a paid individual plan.
- A free plan with no end date
- Bring your own API key
- An annual price lower than the monthly price
- Free plan
- Free: $0 per month, with "Limited AI usage": limited agent requests and limited Chat requests. The page states no end date for the plan.The page is undated. The Free card also lists X features (post, reply and follow).Verified · Nov 10, 2026
- Lowest-priced paid plan for one person
- Plus: $8 per month (USD, US site, 1 user, excluding taxes), which the page calls "our introductory plan": a shared AI usage pool and Grok and Cursor models only, with no third-party models and no own API keys. The docs show annual billing at $80 per year for Plus. Premium, the lowest-priced plan with cloud agents and third-party models, is $30 per month ($300 per year on annual billing).Prices exclude applicable taxes; third-party model usage is not included in the monthly prices and is purchased separately. The annual prices are from https://cursor.com/docs/models-and-pricing, not the pricing page.Verified · Nov 10, 2026
- How usage is priced
- A flat tier per person per month (Plus $8, Premium $30, Super $100, Ultra $200) with a weekly-refreshing usage pool measured as multiples of Plus (Premium 4x, Super 15x, Ultra 50x). On Premium and above, third-party models are billed at API list price from prepaid credit. The docs say Cursor is sold as part of The X Subscription, one subscription for Grok, Cursor and X, and the plan cards list X features. Legacy plans (Hobby, Start, Pro, Pro+, Ultra, Teams Standard and Premium) continue for existing customers.Subscriptions are sold directly through the entity named on the pricing page, with no resellers authorized, per the page FAQ.Verified · Nov 10, 2026
What isn't clear from public sources
The pricing page shows monthly prices only; the annual prices are in the docs. Usage is stated as multiples of Plus ("4x", "15x", "50x"), not as a number of requests or dollars, and third-party model use is billed separately. The plan cards list X features alongside Cursor features, and the docs say legacy plans continue for existing customers. Plus is labelled "our introductory plan" with no other regular price shown.
Zed
Online
Zed lists Personal at $0 (use your own API keys or external agents), Pro at $10 per month with $5 of token usage and edit predictions, and Business at $30 per seat per month. It is its own code editor for macOS, Linux and Windows, with a native agent, external agents and bring-your-own-key.
Our read Relevant if you want a separate code editor with a free plan that works with your own API keys and a low-priced hosted plan.
- A free plan with no end date
- Bring your own API key
- Free plan
- Zed states the Personal plan is "$0 forever": the editor with "unlimited use with your API keys or external agents like Claude Agent, Codex CLI, and more". Zed-hosted models start with Pro; a 14-day Pro trial with $5 of GPT-6 Luna usage is offered with no credit card.The page is undated; the footer shows 2026. The trial ends when the $5 is used or after 14 days.Provider claim · Nov 10, 2026
- Lowest-priced paid plan for one person
- Pro: $10 per month (USD, US site, 1 user), listing edit predictions and $5 of tokens each month. Usage beyond $5 is billed at API list price plus 10%, charged at the end of the month or for every additional $10, whichever comes first, with an optional spending cap. No annual price is shown.The FAQ says Zed Pro bills at API list price plus 10% for hosted usage.Verified · Nov 10, 2026
- How usage is priced
- A flat price per person per month (Pro $10) that includes $5 of token usage, with token-based billing beyond that at API list price plus 10%. Business is $30 per seat per month and bundles no fixed model credit: you use Zed-hosted AI at standard usage rates or bring your own provider keys, in which case spend goes to the provider.Business pricing detail is from the page FAQ.Verified · Nov 10, 2026
What isn't clear from public sources
The pricing page shows monthly prices only. Hosted model usage beyond the $5 included on Pro is billed at API list price plus 10%. Zed Business says SSO, SAML and SCIM are planned but not currently available. The pages read name no support channel by plan.
Warp
Online
Warp lists a Free plan, Build at $20 per month ($18 per month on annual billing) with $20 of agent usage at API rates, Max at $200 per month, and Business at $50 per user per month ($45 annual) for up to 25 seats. It is a terminal app for macOS, Windows and Linux with a coding agent, and can use your own AI inference.
Our read Relevant if you want a terminal-based coding agent, usage priced at API rates and the option to bring your own AI inference.
- A free plan with no end date
- Bring your own API key
- An annual price lower than the monthly price
- Free plan
- Free: $0 per month, with the terminal, Warp Agent CLI access, additional agent usage at pay-as-you-go rates, bring your own AI inference, limited cloud agents access, limited Warp Drive and collaboration, limited cloud conversation storage and individually configured data controls.The page is undated. Warp's privacy docs say telemetry must be enabled to use AI features on the Free plan.Verified · Nov 10, 2026
- Lowest-priced paid plan for one person
- Build: $20 per month billed monthly, or $18 per month billed annually (USD, US site, 1 user), including $20 of agent usage each month at API rates, full Warp Agent access and private email support. Max is $200 per month ($180 annual) with $200 of usage.Monthly and annual prices were read with the page's Monthly and Annual switch, which only changes the displayed prices.Verified · Nov 10, 2026
- How usage is priced
- A price per person per month that includes a dollar amount of agent usage at API rates (Build $20, Max $200); add-on agent usage, auto-reload and a team-wide spend cap are available. Business is $50 per user per month ($45 annual) with $20 of usage per seat.The page footnotes are not shown in the text read.Verified · Nov 10, 2026
What isn't clear from public sources
Included usage is stated in dollars at API rates, so how many tasks it covers depends on the model chosen. The Free plan card marks "Bring your own AI inference" with an asterisk whose footnote was not shown in the page text read. Warp's privacy docs say telemetry must be enabled to use AI features on the Free plan.
OpenAI Codex
Online
OpenAI Codex is included in ChatGPT plans: Free at $0, Go at $8, Plus at $20 and Pro from $100 per month, and Business seats at $20 per month on annual billing ($25 monthly). It runs in a desktop app, a CLI, IDE extensions for VS Code, Xcode and JetBrains, and a cloud, and can be used with a ChatGPT plan or an API key.
Our read Relevant if you already use or would use a ChatGPT plan for coding, or want to pay through an OpenAI API key instead.
- A free plan with no end date
- Bring your own API key
- Runs in the editor you already use
- An annual price lower than the monthly price
- Free plan
- Codex is included in the ChatGPT Free plan ($0 per month, no end date stated), which the page describes as "Test Codex Capabilities" with limited access to flagship models and limited messages and uploads.The page says "Codex is included in your ChatGPT Free, Go, Plus, Pro, Business, or Enterprise plan". It is undated; the footer shows 2015 to 2026.Verified · Nov 10, 2026
- Lowest-priced paid plan for one person
- ChatGPT Go (includes Codex): $8 per month (USD, US site, 1 user), listed with "Test Codex Capabilities". Plus: $20 per month with "Expanded Codex usage", the Codex app for macOS and Windows, and plugins such as Slack, GitHub and Figma. Pro: from $100 per month with "Maximum Codex tasks". Business Standard seat: $20 per month billed annually ($25 billed monthly).Codex is not sold on its own: the price is the ChatGPT plan that includes it.Verified · Nov 10, 2026
- How usage is priced
- A price per person per month for the ChatGPT plan (Free $0, Go $8, Plus $20, Pro from $100), or per seat for Business (Standard $20, Premium $100 per month billed annually; $25 and $125 billed monthly). Local messages and cloud chats share the plan's usage allowance and weekly limits may also apply; credits extend use beyond the plan, and signing in with an API key bills usage at standard API rates instead of plan credits.Credit and limit wording is from the page FAQ. The API-key billing statement is from https://learn.chatgpt.com/docs/auth.Verified · Nov 10, 2026
What isn't clear from public sources
The row prices the coding product as the ChatGPT plans that include it; those plans also cover chat, which another page on this site covers. The usage table gives ranges per model but its text does not label the unit or time window, and plan usage is described as expanded, maximum or limited rather than as a number. The pages read do not state the training default for individual plans.
Claude Code
Online
Claude Code is included in the paid Claude plans: Pro at $17 per month on annual billing ($20 monthly), Max at $100 and Ultra at $200 per month, and Team seats at $20 per month on annual billing ($25 monthly). The Free Claude plan does not include it. It runs in a terminal, VS Code, JetBrains IDEs, a desktop app and the web.
Our read Relevant if you already use or would use a Claude subscription for coding, want VS Code or JetBrains plugins, or want to pay for API usage through your own Console account.
- Bring your own API key
- Runs in the editor you already use
- An annual price lower than the monthly price
- Free plan
- No free plan includes Claude Code. The pricing page lists Claude Code as "No" on the Free plan (which covers chat) and "Claude Code included" on Pro, Max and Ultra.The page is undated; the footer shows 2026.Verified · Nov 10, 2026
- Lowest-priced paid plan for one person
- Pro (includes Claude Code): $17 per month with the annual subscription discount ($200 billed up front), or $20 per month if billed monthly (USD, US site, 1 user, excluding tax). The page lists "More usage than Free, every 5 hours and every week" and gives no number.Max is $100 per month and Ultra $200 per month. Anthropic notes that price and plans are subject to change at its discretion.Verified · Nov 10, 2026
- How usage is priced
- A flat price per person per month with one shared usage limit: a rolling five-hour window plus weekly limits, drawn on by chat on web, desktop and mobile and by Claude Code together. For heavy coding sessions the FAQ says you can switch to pay-as-you-go API credits through a Console account. Team plans are priced per seat (Standard $20, Premium $100 per seat per month on annual billing).Annual plans are billed once up front for the year; the FAQ says annual subscriptions exist for Pro and Team.Verified · Nov 10, 2026
What isn't clear from public sources
The row prices the coding product as the plans that include it; those plans also cover chat, which another page on this site covers. Pro usage is stated only as "more usage than Free" with no number, and Claude Code shares one usage pool with chat. The product page and the pricing page were read together; the product page shows Team at $20 per seat (annual billing) and the pricing page shows the same Standard seat plus a $100 Premium seat.
GitHub Copilot
Online
GitHub Copilot lists Free at $0, Pro at $10, Pro+ at $39 and Max at $100 per user per month, plus Business at $19 and Enterprise at $39 per user per month. AI usage is paid in GitHub AI Credits ($0.01 each) with a monthly allowance in each plan. It names plugins for VS Code, Visual Studio, JetBrains IDEs, Xcode, Neovim, Eclipse and Zed.
Our read Relevant if you want a free tier, per-user prices from $10 and plugins for the editor you already use.
- A free plan with no end date
- Bring your own API key
- Runs in the editor you already use
- Free plan
- Free: $0 USD, no credit card required, with 2,000 completions per month, access to Haiku 4.5, GPT-5 mini and more, Copilot CLI and Community Support. The page also lists limited chat and agent usage and supports the CLI and agent mode on Free.Verified students are pointed to a separate Copilot Student plan. The page is undated; the footer shows 2026.Verified · Nov 10, 2026
- Lowest-priced paid plan for one person
- Pro: $10 per user per month (USD, US site, 1 user, billed monthly; the page shows no annual price). It includes Copilot in the cloud and code review, access to third-party agents (Claude Code and Codex), model selection and $15 of monthly AI credits ($10 base plus $5 flex).The page says flex allotments may change over time. Pro+ is $39 and Max $100 per user per month.Verified · Nov 10, 2026
- How usage is priced
- A flat price per user per month by plan (Free $0, Pro $10, Pro+ $39, Max $100; Business $19 and Enterprise $39 per user per month). AI usage is paid in GitHub AI Credits, where 1 credit = $0.01 and every plan includes a monthly allowance; agent mode, code review, Copilot in the cloud, the CLI and the app use credits. Past the allowance you can wait for the next cycle, set a dollar budget for more credits at $0.01 each (paid plans), or switch to a cheaper model.Credit rules are from the page FAQ.Verified · Nov 10, 2026
What isn't clear from public sources
The plans page shows monthly prices only. Credits per task depend on the model and the page gives no per-task figure. The page says flex allotments "may change over time" and that model options vary by feature. Third-party agent delegation is listed for Pro+ and Max in the comparison table although the Pro card lists access to Claude Code and Codex.
Kiro
Online
Kiro lists a Free tier with 50 credits per month, Pro at $20 per user per month with 1,000 credits, Pro+ at $40, Pro Max at $100 and Power at $200, with add-on credits at $0.04 each. Teams use credit pooling at $20 per seat per month with 500 credits per seat, for pools of 2 or more seats.
Our read Relevant if you want a free tier, a credit-based plan and Kiro's own IDE, CLI or web app.
- A free plan with no end date
- Free plan
- Kiro Free: $0 per month with 50 credits per month and access to open weight models and Claude Sonnet 4.5, with limits. The FAQ calls it the "perpetual Kiro Free tier". It is not available in Kiro Enterprise or AWS GovCloud (US) Regions.The page is undated; the footer lists legal links without a date.Verified · Nov 10, 2026
- Lowest-priced paid plan for one person
- Kiro Pro: $20 per user per month (USD, 1 user) with 1,000 credits. When you first upgrade from Free to Pro, the rest of that month is free and you get 500 trial credits to use within 12 months; upgrading requires a valid credit card. No annual price is shown. Pro+ is $40 (2,000 credits), Pro Max $100 (5,000) and Power $200 (10,000) per user per month.Plan fees and credits are prorated to the second, per the page FAQ.Verified · Nov 10, 2026
- How usage is priced
- A price per user per month that includes a monthly credit allowance. A credit is a unit of work in response to prompts: simple prompts can use less than 1 credit, complex prompts such as executing a spec task usually more, and different models use credits at different rates. Unused plan credits do not roll over. Add-on credits cost $0.04 each (packs from $5 for 125 credits up to $100) and expire 12 months after purchase. Teams use credit pooling instead of individual tiers.Credit rules are from the page FAQ.Verified · Nov 10, 2026
What isn't clear from public sources
The pricing page shows monthly prices only and no bring-your-own-key option. A credit is a unit of work whose cost varies by prompt and model, and the page gives no per-task figure. Kiro Enterprise is purchased and billed through AWS with no free tier. Support terms by plan are not stated.
JetBrains AI
Online
JetBrains AI lists AI Free with 3 AI credits every 30 days, AI Pro at $10 per user per month ($100 per year) with 10 credits, and AI Ultimate at $30 per month ($300 per year) with 35 credits, where 1 credit equals USD 1.00 of AI use. Organizations buy credits and can add Air Governance plans, with Business listed at $20 per seat per month.
Our read Relevant if you work in JetBrains IDEs and want monthly or yearly billing, usage priced at the model providers' API rates, or your own API key.
- A free plan with no end date
- Bring your own API key
- An annual price lower than the monthly price
- Free plan
- JetBrains AI states AI Free includes unlimited AI completion in JetBrains IDEs and 3 AI credits every 30 days, with no end date stated and no price. The page says it is only available for commercial IDE versions and the FAQ says it is not available in IntelliJ IDEA without an Ultimate subscription, in PyCharm without Pro or in Android Studio. A free 30-day AI Pro trial is offered when you first activate JetBrains AI in a JetBrains IDE.The buy page redirects to https://www.jetbrains.com/store/ai/ and is undated; the footer shows 2026.Provider claim · Nov 10, 2026
- Lowest-priced paid plan for one person
- AI Pro: $10.00 per user per month on monthly billing, or $100.00 per user per year on yearly billing (the page says yearly billing saves 2 months); USD, 1 user, with 10 AI credits every 30 days and access to the Junie, Claude Agent, Codex and Gemini agents. The page says taxes may be added and prices shown are approximate.Both figures were read with the page's billing toggle, which only changes the displayed prices. AI Ultimate is $30 per month or $300 per year with 35 credits.Verified · Nov 10, 2026
- How usage is priced
- A price per user (monthly or yearly) that includes a set number of AI credits, renewed every 30 days from first activation with no roll-over. One AI credit equals USD 1.00 of AI use, and third-party model usage is billed at public API rates with no surcharge. AI Pro and AI Ultimate users can buy top-up credits valid for 12 months. Organizations have no fixed credit plans and buy credits in any amount for shared use.Credit rules are from the page FAQ.Verified · Nov 10, 2026
What isn't clear from public sources
The buy page redirects to a store page and lists prices as approximate. It shows a $20 per seat per month price for the Air Governance Business plan with a "Get in touch" button, while the FAQ says Business is free during early access; usage is paid separately in both cases. AI Free is not available in every IDE edition, and the pages name no plugin for non-JetBrains editors.
0 selected:
Full comparison table
| Name | Free plan | Lowest-priced paid plan for one person | How usage is priced | Usage included | Where it runs | Models and bring your own key | Agent features | Team plan and admin controls | Your code and data (provider statements) | Support |
|---|---|---|---|---|---|---|---|---|---|---|
| Amazon Q Developer | A Free Tier is listed: "Advanced capabilities at zero cost", with 50 agentic requests per month (Q&A chat and agentic coding in the IDE plugins and the CLI) and Java upgrade transformation of up to 1,000 lines of code per month. The page says the Free Tier is perpetual with monthly limits, for AWS IAM users or AWS Builder ID users, and that Free Tier limits in the IDE are for Builder ID users only.The page is undated. The Free Tier summary also lists "Access latest Claude models" and use in the IDE or CLI.Verified · Nov 10, 2026 | Pro Tier: $19 per user per month (USD, US site, 1 user), billed monthly as a subscription that starts when the user first uses a Pro feature. No annual price is shown. It lists increased agentic request limits ("Included (with limits)"), Java transformation of 4,000 lines of code per user per month pooled at the AWS payer-account level (extra lines $0.003 per line), IAM Identity Center support with admin dashboards and controls, and IP indemnity.The page FAQ says the monthly fee is charged per user and shown daily as a fraction of a month, for example $19 x 1/30 of a month = about $0.63 a day.Verified · Nov 10, 2026 | Per user per month ($19) for Pro, charged only for each user who activates it; the Free Tier has no fee. Usage is counted in agentic requests (any Q&A chat or agentic coding interaction through the IDE or the CLI) and, for Java upgrades, in lines of code submitted ($0.003 per extra line on Pro).Billing detail is from the page FAQ on activating and cancelling a Pro subscription.Verified · Nov 10, 2026 | Free: 50 agentic requests per month and 1,000 lines of code per month for Java upgrades. Pro: "Included (with limits)" agentic requests (the page gives no number) and 4,000 lines of code per user per month for Java upgrades, pooled at the payer-account level.The page says usage for an account may be changed depending on regional factors, payment history or approved quota increases.Verified · Nov 10, 2026 | Plugins for JetBrains IDEs, VS Code, Visual Studio and Eclipse (preview), and a command line tool for terminals (locally and over SSH); the AWS Management Console, Microsoft Teams and Slack for AWS questions. A notice on the page says AWS will discontinue support for Amazon Q Developer IDE plugins on April 30, 2027 and suggests Kiro for similar capabilities.The FAQ at https://aws.amazon.com/q/developer/faqs/ also answers that a Q Developer Pro subscription can be used with the Kiro IDE or Kiro CLI, using agentic request inference calls.Verified · Nov 10, 2026 | The pricing page lists "Access latest Claude models" under the Free Tier summary. The FAQ says Amazon Q Developer is powered by Amazon Bedrock and uses cross-Region inference. The pages read state no bring-your-own-key option.FAQ wording is from https://aws.amazon.com/q/developer/faqs/. Neither page is dated.Verified · Nov 10, 2026 | Agentic coding in the IDE plugins and CLI, inline code suggestions, inline chat, vulnerability scanning, CLI completions, and agents for code transformation (Java upgrades; the pricing page also lists .NET app transformation for Pro).Feature wording is from the product and pricing pages.Verified · Nov 10, 2026 | The Pro Tier table lists an admin dashboard with user and policy management ("No" for Free, "Yes" for Pro), IAM Identity Center support with admin controls, and IP indemnity. The page says members with subscriptions linked to an organization with IAM Identity Center get Pro limits at the user level. No separate team plan or team price is listed.Only the per-user Pro price is shown for organizations.Verified · Nov 10, 2026 | Amazon Q Developer states that Amazon Q Developer Pro does not use your content for service improvement and that, for Pro Tier users, content is not used to train any underlying foundation models; it states the Free Tier might use certain content (questions, responses and code) for service improvement, including model training, with opt-out in IDE settings or through an AI services opt-out policy in AWS Organizations. The pricing table lists data collection as "Opt-out available" for Free and "Automatically opted out" for Pro.FAQ page is undated. The pricing-table row is from https://aws.amazon.com/q/developer/pricing/.Provider claim · Nov 10, 2026 | Not publicly statedThe pricing, product and FAQ pages read name no support channel or response time for the Free or Pro Tier. |
| Cursor | Free: $0 per month, with "Limited AI usage": limited agent requests and limited Chat requests. The page states no end date for the plan.The page is undated. The Free card also lists X features (post, reply and follow).Verified · Nov 10, 2026 | Plus: $8 per month (USD, US site, 1 user, excluding taxes), which the page calls "our introductory plan": a shared AI usage pool and Grok and Cursor models only, with no third-party models and no own API keys. The docs show annual billing at $80 per year for Plus. Premium, the lowest-priced plan with cloud agents and third-party models, is $30 per month ($300 per year on annual billing).Prices exclude applicable taxes; third-party model usage is not included in the monthly prices and is purchased separately. The annual prices are from https://cursor.com/docs/models-and-pricing, not the pricing page.Verified · Nov 10, 2026 | A flat tier per person per month (Plus $8, Premium $30, Super $100, Ultra $200) with a weekly-refreshing usage pool measured as multiples of Plus (Premium 4x, Super 15x, Ultra 50x). On Premium and above, third-party models are billed at API list price from prepaid credit. The docs say Cursor is sold as part of The X Subscription, one subscription for Grok, Cursor and X, and the plan cards list X features. Legacy plans (Hobby, Start, Pro, Pro+, Ultra, Teams Standard and Premium) continue for existing customers.Subscriptions are sold directly through the entity named on the pricing page, with no resellers authorized, per the page FAQ.Verified · Nov 10, 2026 | Free: limited agent and Chat requests (no number). Plus: a shared usage pool shown as "Base". Premium 4x Plus, Super 15x Plus, Ultra 50x Plus; the allowance refreshes weekly. Teams: 4x to 50x Plus by seat tier, pooled on Enterprise. Third-party model usage is not part of the included pool.The page FAQ says every tier includes a pool of AI usage shared across Cursor and Grok, with top-offs available when it runs out.Verified · Nov 10, 2026 | The Cursor desktop app for macOS, Windows and Linux, Cursor on the web, the Cursor CLI and a mobile app. The pages read name no plugin for other editors.The page is undated; the surfaces are listed on the download page and the site navigation.Verified · Nov 10, 2026 | Plus: Cursor's own models (Grok 4.7, 4.6, 4.5 and Composer 2.5) at fixed settings. Premium and above: third-party models (the docs name OpenAI, Anthropic, Google, SpaceXSI and more) at API list price from prepaid credit. Your own API key for OpenAI, Anthropic, Google, Azure OpenAI or AWS Bedrock works with chat models on the individual Premium, Super and Ultra plans (Plus excludes own keys); Tab completion keeps using Cursor's own models.Bring-your-own-key detail is from https://cursor.com/help/models-and-usage/api-keys. On Teams and Enterprise a Cursor Token Rate of $0.25 per million tokens applies to eligible third-party model requests, including those made with your own key.Verified · Nov 10, 2026 | Agent on every plan; cloud agents and automations, code hosting via Origin, and AI code reviews via Bugbot on Premium and above; a CLI; Grok Bot on Premium and above.Feature rows are from the pricing-page comparison table.Verified · Nov 10, 2026 | Teams: $30+ per user per month on the pricing page (the docs show $30, $100 or $200 per seat per month by seat tier, mix and match allowed; Plus is not available for teams). It lists centralized billing and administration, usage analytics, team-wide privacy settings, custom chat data retention and SAML/OIDC SSO. Enterprise (custom price) adds invoice and PO billing, SCIM, granular spend controls, audit logs and service accounts, and priority support.Seat-tier detail is from https://cursor.com/docs/models-and-pricing.Verified · Nov 10, 2026 | Cursor states that with Privacy Mode on, customer data is not used for training by Cursor and it keeps zero data retention agreements with all providers, though providers may run abuse classifiers; with Privacy Mode off, it may use and store codebase data, prompts, editor actions and code snippets to improve its features and train its models. It states Zero Data Retention does not apply when you use your own API keys and that requests still go through its backend. Its security page states AIUC-1, ISO/IEC 27001:2022, ISO/IEC 42001:2023 and SOC 2 Type II.The data-use page shows "Last updated September 3, 2026". The security wording is from https://cursor.com/security (undated).Provider claim · Nov 10, 2026 | The pricing comparison lists priority support for Enterprise only. The site links Docs, a Help center, a Forum and Contact Support; no response times are stated.The page is undated.Verified · Nov 10, 2026 |
| Zed | Zed states the Personal plan is "$0 forever": the editor with "unlimited use with your API keys or external agents like Claude Agent, Codex CLI, and more". Zed-hosted models start with Pro; a 14-day Pro trial with $5 of GPT-6 Luna usage is offered with no credit card.The page is undated; the footer shows 2026. The trial ends when the $5 is used or after 14 days.Provider claim · Nov 10, 2026 | Pro: $10 per month (USD, US site, 1 user), listing edit predictions and $5 of tokens each month. Usage beyond $5 is billed at API list price plus 10%, charged at the end of the month or for every additional $10, whichever comes first, with an optional spending cap. No annual price is shown.The FAQ says Zed Pro bills at API list price plus 10% for hosted usage.Verified · Nov 10, 2026 | A flat price per person per month (Pro $10) that includes $5 of token usage, with token-based billing beyond that at API list price plus 10%. Business is $30 per seat per month and bundles no fixed model credit: you use Zed-hosted AI at standard usage rates or bring your own provider keys, in which case spend goes to the provider.Business pricing detail is from the page FAQ.Verified · Nov 10, 2026 | Zed states Personal has unlimited use with your own API keys or external agents; Pro includes unlimited edit predictions and $5 of tokens per month; Business includes unlimited edit predictions and no fixed allotment of model credit.The page also says the Pro trial balance of $5 is shared across Zed and Delta.Provider claim · Nov 10, 2026 | Zed, its own code editor, installed on macOS, Linux and Windows (the download page showed version 1.27.0, dated November 04, 2026). The AI docs describe the Agent Panel, parallel agents, external ACP agents and terminal threads inside Zed.The download page is https://zed.dev/download. The pages read name no plugin for other editors.Verified · Nov 10, 2026 | Zed-hosted models from Anthropic, OpenAI, Google and others at provider price plus 10% (the docs list, for example, Claude Opus 5.5 and Claude Opus 5); your own keys for Amazon Bedrock, Anthropic, GitHub Copilot, DeepSeek, Google AI, LM Studio, Mistral, Ollama, OpenAI, OpenRouter, Vercel and more; and external agents such as Claude Agent and Codex CLI. The Pro trial includes GPT-6 Luna only.Bring-your-own-key providers are from the pricing FAQ at https://zed.dev/pricing.Verified · Nov 10, 2026 | The Agent Panel, parallel agents across projects and worktrees, Zed's native agent (with built-in tools, profiles, skills, instructions and MCP servers), external ACP agents, terminal threads for agent CLIs, an inline assistant, edit prediction and Git commit message generation.Feature wording is from the AI overview docs.Verified · Nov 10, 2026 | Business: $30 per seat per month, self-serve with no minimum seat count (order forms at 25+ seats), listing org-wide AI model policies, data governance controls, unified spend visibility, role-based access controls and edit predictions. Admins can restrict or disable Zed-hosted models by model, turn off edit predictions org-wide and lock data-sharing settings. SSO, SAML and SCIM are planned but not currently available; Business has no free trial.Month-to-month plans are available to everyone, per the FAQ.Verified · Nov 10, 2026 | Zed states it does not retain your prompts or code context by default; that for Zed-hosted models it has no-training commitments from model providers and provider agreements require zero data retention for inference requests, except for provider-designated models with safety retention (such as Anthropic's Covered Models); and that it retains AI data only when you explicitly share feedback or opt in to training data collection.The page is undated. Agents and edit prediction providers you choose follow their own data terms, per the same page.Provider claim · Nov 10, 2026 | Not publicly statedThe pricing page names a sales email address for order forms at 25+ seats and larger teams but states no support channel or response time for the Personal, Pro or Business plans. |
| Warp | Free: $0 per month, with the terminal, Warp Agent CLI access, additional agent usage at pay-as-you-go rates, bring your own AI inference, limited cloud agents access, limited Warp Drive and collaboration, limited cloud conversation storage and individually configured data controls.The page is undated. Warp's privacy docs say telemetry must be enabled to use AI features on the Free plan.Verified · Nov 10, 2026 | Build: $20 per month billed monthly, or $18 per month billed annually (USD, US site, 1 user), including $20 of agent usage each month at API rates, full Warp Agent access and private email support. Max is $200 per month ($180 annual) with $200 of usage.Monthly and annual prices were read with the page's Monthly and Annual switch, which only changes the displayed prices.Verified · Nov 10, 2026 | A price per person per month that includes a dollar amount of agent usage at API rates (Build $20, Max $200); add-on agent usage, auto-reload and a team-wide spend cap are available. Business is $50 per user per month ($45 annual) with $20 of usage per seat.The page footnotes are not shown in the text read.Verified · Nov 10, 2026 | Free: pay-as-you-go agent usage and limited cloud agents. Build: $20 of agent usage per month at API rates and extended cloud agents access. Max: $200 of usage per month (10x Build). Business: $20 of usage per seat per month.Highest codebase indexing limits are listed from Build.Verified · Nov 10, 2026 | The Warp terminal app for macOS, Windows and Linux and the Warp Agent CLI, which the page describes as a coding agent that works in any terminal; cloud agents run in the cloud.Platforms are from the download page. It also lists Warp Preview and a "request early access" product called Warp Factories.Verified · Nov 10, 2026 | Build lists full Warp Agent access with "frontier and open-source models from OpenAI, Anthropic, z.ai, and more". Free lists bring your own AI inference; Business lists bring your own API keys and custom inference endpoints; Enterprise lists routing inference through your own cloud.Specific model names are not listed on the pricing page.Verified · Nov 10, 2026 | Warp Agent in the terminal and as a CLI, cloud agents (limited on Free, extended on Build) and a "use any harness in the cloud (beta)" option, codebase indexing, and Warp Drive objects and collaboration.Feature wording is from the pricing page.Verified · Nov 10, 2026 | Business: $50 per user per month ($45 annual) for up to 25 seats, with team usage metrics, admin-configurable data controls, bring your own API keys and custom inference endpoints, and SAML-based SSO. Enterprise (custom): no seat limit stated, custom shared usage pools, advanced spend controls, enterprise admin controls and data governance, an analytics API, multi-admin support, self-hosted cloud agents and a dedicated account manager.Plan lists are from the pricing page.Verified · Nov 10, 2026 | Warp states that Business and Enterprise plans are covered by its Zero Data Retention agreement and that no AI interaction or console data is collected for them; for other plans it may collect AI interactions and console inputs unless you opt out of "Help improve Warp" (paid plans can opt out and continue using AI; the Free plan needs telemetry on for AI features); and it states selling usage data will never be part of its business model.The statements are as stated on the privacy docs page.Provider claim · Nov 10, 2026 | Build lists private email support; Enterprise lists a dedicated account manager and a shared Slack channel. Support for the Free and Business plans is not stated.The page is undated.Verified · Nov 10, 2026 |
| OpenAI Codex | Codex is included in the ChatGPT Free plan ($0 per month, no end date stated), which the page describes as "Test Codex Capabilities" with limited access to flagship models and limited messages and uploads.The page says "Codex is included in your ChatGPT Free, Go, Plus, Pro, Business, or Enterprise plan". It is undated; the footer shows 2015 to 2026.Verified · Nov 10, 2026 | ChatGPT Go (includes Codex): $8 per month (USD, US site, 1 user), listed with "Test Codex Capabilities". Plus: $20 per month with "Expanded Codex usage", the Codex app for macOS and Windows, and plugins such as Slack, GitHub and Figma. Pro: from $100 per month with "Maximum Codex tasks". Business Standard seat: $20 per month billed annually ($25 billed monthly).Codex is not sold on its own: the price is the ChatGPT plan that includes it.Verified · Nov 10, 2026 | A price per person per month for the ChatGPT plan (Free $0, Go $8, Plus $20, Pro from $100), or per seat for Business (Standard $20, Premium $100 per month billed annually; $25 and $125 billed monthly). Local messages and cloud chats share the plan's usage allowance and weekly limits may also apply; credits extend use beyond the plan, and signing in with an API key bills usage at standard API rates instead of plan credits.Credit and limit wording is from the page FAQ. The API-key billing statement is from https://learn.chatgpt.com/docs/auth.Verified · Nov 10, 2026 | The usage table gives ranges per model for Plus and for Standard Business: GPT-6.1 Sol 15-160, GPT-6 Astra 5-45, GPT-6 Sol 15-150, GPT-6 Luna 350-3,000, GPT-5.6 Sol 10-100, GPT-5.6 Luna 250-2,000 and GPT-5.5 15-80, with the API Key column "usage-based". Free and Go are described as limited and Pro as "Maximum Codex tasks".The table text does not label the unit or time window of the ranges. Enterprise and Edu plans without flexible pricing have the same per-seat limits as Plus for most features, per the page.Verified · Nov 10, 2026 | The Codex app (macOS and Windows), the Codex CLI (macOS, Linux and Windows), an IDE extension (VS Code and compatible editors, Xcode and JetBrains IDEs), Codex cloud (requires signing in with ChatGPT) and a mobile app that connects to a computer.Surfaces are from https://learn.chatgpt.com/docs/codex/ide, https://learn.chatgpt.com/docs/codex/cli and https://learn.chatgpt.com/docs/auth. The pricing page names the app for macOS and Windows.Verified · Nov 10, 2026 | The usage table lists GPT-6.1 Sol, GPT-6 Astra, GPT-6 Sol, GPT-6 Luna, GPT-5.6 Sol, GPT-5.6 Luna and GPT-5.5. The auth page says you can sign in with ChatGPT for subscription access or with an API key for usage-based access (from the OpenAI dashboard), in the desktop app, CLI and IDE extension; some features need ChatGPT sign-in and Codex cloud requires it.API-key detail is from https://learn.chatgpt.com/docs/auth.Verified · Nov 10, 2026 | Coding tasks run locally in the app, CLI or IDE extension and in Codex cloud, with plugins that connect tools such as Slack, GitHub and Figma; the docs index lists approvals and sandboxing, auto-review, scheduled automations and Codex Security.The docs index is https://learn.chatgpt.com/llms.txt.Verified · Nov 10, 2026 | Business, for teams of 2 to 200: Standard seat $20 and Premium seat $100 per month billed annually ($25 and $125 monthly), including ChatGPT, ChatGPT Work and Codex, central billing and administration, usage analytics, budgeting and spend controls, SAML SSO and MFA. Enterprise (custom price) lists a choice of pay-as-you-go or fixed seat pricing, SCIM, EKM, domain verification, RBAC, audit logs and the Compliance API.Read with the page's Business and Enterprise tab, which only changes the displayed content.Verified · Nov 10, 2026 | OpenAI Codex states that the Business plan has "No training on your business data by default". The pages read do not state the training default for the Free, Go, Plus or Pro plans. The auth docs state that with ChatGPT sign-in usage follows the workspace's ChatGPT retention and residency settings, and with an API key it follows the API organization's retention and data-sharing settings.Business wording is from the pricing page; the auth wording is from https://learn.chatgpt.com/docs/auth.Provider claim · Nov 10, 2026 | Not publicly statedThe pricing page and the Codex docs pages read name no support channel or response time for any plan. |
| Claude Code | No free plan includes Claude Code. The pricing page lists Claude Code as "No" on the Free plan (which covers chat) and "Claude Code included" on Pro, Max and Ultra.The page is undated; the footer shows 2026.Verified · Nov 10, 2026 | Pro (includes Claude Code): $17 per month with the annual subscription discount ($200 billed up front), or $20 per month if billed monthly (USD, US site, 1 user, excluding tax). The page lists "More usage than Free, every 5 hours and every week" and gives no number.Max is $100 per month and Ultra $200 per month. Anthropic notes that price and plans are subject to change at its discretion.Verified · Nov 10, 2026 | A flat price per person per month with one shared usage limit: a rolling five-hour window plus weekly limits, drawn on by chat on web, desktop and mobile and by Claude Code together. For heavy coding sessions the FAQ says you can switch to pay-as-you-go API credits through a Console account. Team plans are priced per seat (Standard $20, Premium $100 per seat per month on annual billing).Annual plans are billed once up front for the year; the FAQ says annual subscriptions exist for Pro and Team.Verified · Nov 10, 2026 | Pro: "more usage than Free" every 5 hours and every week (no number given). Max: 5x Pro usage every 5 hours and every week, plus $100 a month in API credits. Ultra: no 5-hour limit for verified accounts, 10x Pro usage every week, plus $200 a month in API credits. Team Standard seat: "more usage than Pro"; Premium seat: 5x more usage than a Standard seat.Usage limits apply and depend on the model, chat length and file size, per the page footnote.Verified · Nov 10, 2026 | The terminal (CLI), a VS Code extension, a JetBrains plugin (it requires the CLI), a desktop app, the web, iOS and Android apps, GitHub and Slack. Available for macOS, Linux and Windows.Surfaces are listed on https://claude.com/product/claude-code. The docs page says most surfaces need a Claude subscription or an Anthropic Console account.Verified · Nov 10, 2026 | The pricing page lists Claude models: Fable (paid with usage credits on Pro; included up to half the weekly limit on Max), Opus, Sonnet and Haiku. The docs say the Terminal CLI, VS Code and JetBrains also support third-party providers, and that most surfaces need a Claude subscription or an Anthropic Console account (API usage billed through that account).The provider-routing detail is from https://code.claude.com/docs/en/overview. Neither page is dated.Verified · Nov 10, 2026 | Handing Claude a bug fix, test or multi-day migration with permissions you set, steered from desktop, terminal, IDE, Slack or the web. The product page lists Code Review (generally available for Team and Enterprise, October 15, 2026 blog item), Claude Code on web and mobile, scheduled tasks, and Projects to group coding sessions and supervise several Claude agents (Claude Code Desktop, September 17, 2026 blog item).Dated items are blog entries listed on the product page.Verified · Nov 10, 2026 | Team plan, 2 to 150 seats: Standard seat $20 and Premium seat $100 per seat per month billed annually ($25 and $125 billed monthly), with Claude Code, central billing and administration, SSO, admin controls for connectors, mix-and-match seat types and up to $500 per month in pooled API credits. Enterprise (self-serve) is listed at $20 per seat per month plus usage at API rates, billed annually, 20 seat minimum, and adds admin spend limits, role-based access, SCIM, audit logs and a Compliance API.The product page shows the Team seat price as $20 per seat per month if billed annually, for teams of 2 to 150.Verified · Nov 10, 2026 | Claude Code states that for Free, Pro, Max and Ultra accounts it lets you choose whether your data is used to improve future Claude models and that it trains new models on data from those accounts when the setting is on, including Claude Code use, with a 5-year retention period if allowed and 30 days if not. It states that under commercial terms (Team, Enterprise, API) it does not train generative models on code or prompts sent to Claude Code unless the customer opts in (for example through the Developer Partner Program), with 30-day standard retention and zero data retention available to qualified Enterprise accounts. The Team plan card lists "No model training on your content by default".The docs page is undated. The Team card wording is from https://claude.com/pricing.Provider claim · Nov 10, 2026 | The pricing comparison lists customer success support at certain spend thresholds for the sales-assisted Enterprise plan only, and the site links a Support center. No support channel or response time is listed for Pro, Max, Ultra or Team.The page is undated.Verified · Nov 10, 2026 |
| GitHub Copilot | Free: $0 USD, no credit card required, with 2,000 completions per month, access to Haiku 4.5, GPT-5 mini and more, Copilot CLI and Community Support. The page also lists limited chat and agent usage and supports the CLI and agent mode on Free.Verified students are pointed to a separate Copilot Student plan. The page is undated; the footer shows 2026.Verified · Nov 10, 2026 | Pro: $10 per user per month (USD, US site, 1 user, billed monthly; the page shows no annual price). It includes Copilot in the cloud and code review, access to third-party agents (Claude Code and Codex), model selection and $15 of monthly AI credits ($10 base plus $5 flex).The page says flex allotments may change over time. Pro+ is $39 and Max $100 per user per month.Verified · Nov 10, 2026 | A flat price per user per month by plan (Free $0, Pro $10, Pro+ $39, Max $100; Business $19 and Enterprise $39 per user per month). AI usage is paid in GitHub AI Credits, where 1 credit = $0.01 and every plan includes a monthly allowance; agent mode, code review, Copilot in the cloud, the CLI and the app use credits. Past the allowance you can wait for the next cycle, set a dollar budget for more credits at $0.01 each (paid plans), or switch to a cheaper model.Credit rules are from the page FAQ.Verified · Nov 10, 2026 | GitHub Copilot states that code completions and next edit suggestions do not use credits and are unlimited on every paid plan; Free has 2,000 completions per month. Monthly AI credits: Pro $15 ($10 base, $5 flex), Pro+ $70 ($39 base, $31 flex), Max $200 ($100 base, $100 flex); Business 1,900 credits per user, pooled org-wide; Enterprise 2,000 more credits per user than Business.Free lists "limited chat and agent usage" with no number. The pooled-usage row lists Business and Enterprise as included and individual plans as not included.Provider claim · Nov 10, 2026 | Editors and IDEs: VS Code, Visual Studio, Xcode, JetBrains IDEs, Neovim, Eclipse, Raycast, SQL Server Management Studio and Zed; also github.com, GitHub Mobile, the Copilot CLI, a Copilot desktop app, Slack and Teams (the FAQ).Agent mode is listed in VS Code, Visual Studio, JetBrains, Eclipse and Xcode.Verified · Nov 10, 2026 | The plan table lists, for each plan, Anthropic Claude models (including Haiku 4.5, Sonnet 5.5 and Opus 5.5), OpenAI GPT models (GPT-5 mini through GPT-6.1), Microsoft MAI-Code-1.1-Flash, Google Gemini models, xAI Grok 4.5 to 4.7 and Kimi K3; the page says model options may vary by feature. GitHub Docs say bring your own key lets you use Copilot with compatible models you or your organization configure, supported in VS Code, JetBrains IDEs, Xcode, the Copilot CLI, the Copilot app and the SDK, billed by your provider; an organization policy can turn it off in IDEs for Business and Enterprise.The bring-your-own-key detail is from https://docs.github.com/en/copilot/concepts/models/bring-your-own-key (undated).Verified · Nov 10, 2026 | Agent mode in VS Code, Visual Studio, JetBrains, Eclipse and Xcode; MCP servers; custom instructions and agents; Copilot in the cloud (assign work and it creates a pull request) and code review on Pro and above; the Copilot CLI on every plan; access to third-party agents (Claude Code and Codex) on Pro and delegating tasks to them (Preview) on Pro+ and Max.Free lists agent mode and the CLI but not Copilot in the cloud or code review.Verified · Nov 10, 2026 | Business: $19 per user per month with 1,900 credits per user pooled org-wide, access control, budget control and governance, web-based support, and the comparison table lists SAML SSO, user management in github.com, usage metrics and enterprise-grade security as included. Enterprise: $39 per user per month with 2,000 more credits per user and priority access to new models. Individual plans list SAML SSO as not included.Both plans also list IP indemnity and data privacy.Verified · Nov 10, 2026 | GitHub Copilot states that for Free, Pro and Pro+ subscribers GitHub may use Copilot interaction data (prompts, suggestions and code snippets) to train and improve AI models, and that individual subscribers can opt out at any time in their Copilot settings. For Business and Enterprise it states that data is used under its Data Protection Agreement to deliver and maintain the service, and the Business card lists data privacy.The statements are in the page FAQ. The page does not name Max in the individual-subscriber answer.Provider claim · Nov 10, 2026 | Free: Community Support. Business: web-based support. Support for Pro, Pro+, Max and Enterprise is not itemized beyond a "Contact sales" link.The page is undated.Verified · Nov 10, 2026 |
| Kiro | Kiro Free: $0 per month with 50 credits per month and access to open weight models and Claude Sonnet 4.5, with limits. The FAQ calls it the "perpetual Kiro Free tier". It is not available in Kiro Enterprise or AWS GovCloud (US) Regions.The page is undated; the footer lists legal links without a date.Verified · Nov 10, 2026 | Kiro Pro: $20 per user per month (USD, 1 user) with 1,000 credits. When you first upgrade from Free to Pro, the rest of that month is free and you get 500 trial credits to use within 12 months; upgrading requires a valid credit card. No annual price is shown. Pro+ is $40 (2,000 credits), Pro Max $100 (5,000) and Power $200 (10,000) per user per month.Plan fees and credits are prorated to the second, per the page FAQ.Verified · Nov 10, 2026 | A price per user per month that includes a monthly credit allowance. A credit is a unit of work in response to prompts: simple prompts can use less than 1 credit, complex prompts such as executing a spec task usually more, and different models use credits at different rates. Unused plan credits do not roll over. Add-on credits cost $0.04 each (packs from $5 for 125 credits up to $100) and expire 12 months after purchase. Teams use credit pooling instead of individual tiers.Credit rules are from the page FAQ.Verified · Nov 10, 2026 | Free 50 credits per month; Pro 1,000; Pro+ 2,000; Pro Max 5,000; Power 10,000 per month, plus 500 trial credits (within 12 months) on the first Pro upgrade. Team credit pooling: 500 included credits per seat per month, pooled across the seats assigned to the pool.Prompts in vibe mode or spec mode, spec refinement, task execution and agent hook execution all use credits, per the FAQ.Verified · Nov 10, 2026 | The Kiro IDE (macOS, Windows and Linux), the Kiro CLI, Kiro Crew, Kiro Web (Pro, Pro+, Pro Max and Power plans) and ACP-compatible IDEs; the downloads page lists a mobile app as "Request early access". The pricing FAQ says subscriptions can also be used for automation such as reviews during CI/CD, and that use through third-party automation harnesses outside Kiro's native interfaces is not permitted.Surfaces are from https://kiro.dev/downloads/ and the pricing FAQ.Verified · Nov 10, 2026 | By default prompts are processed by Auto, which uses a mix of frontier models plus specialized models; you can also choose OpenAI GPT-5.6 (Sol, Terra and Luna), Anthropic Claude Opus 5, Opus 4.8, Sonnet 5, Sonnet 4.6, Sonnet 4.5 and Haiku 4.5, and open weight models. The pages read state no bring-your-own-key option.Model availability can vary by country or region, per the FAQ.Verified · Nov 10, 2026 | Vibe mode and spec mode, spec refinement and task execution, agent hooks, workflows and Kiro powers (no additional charge), Kiro Web running on cloud compute with no separate charge, and Kiro Crew.Feature wording is from the pricing FAQ.Verified · Nov 10, 2026 | Credit pooling for teams of 2 or more: admins buy seats at $20 per seat per month, each seat adds 500 credits to a shared pool, add-on credits are $0.04 each, and admins can create groups, set usage limits for individuals, groups and profiles and watch dashboards with alerts at 80% and 100%. The page lists centralized billing, SSO and enterprise security controls; enterprise plans are bought and billed through AWS and sign in through AWS IAM Identity Center or an external identity provider.Enterprise sign-in detail is from the pricing FAQ.Verified · Nov 10, 2026 | Kiro states that for Free Tier users and paid individual subscribers, content may be used for service improvement subject to opt-out (and Free Tier inputs may be stored for up to 60 days for abuse detection), stored in the US East (N. Virginia) Region; for enterprise users it states content is not used for service improvement.The page is undated.Provider claim · Nov 10, 2026 | The pricing page links Billing support, and the site footer lists Docs, FAQs, Discord and bug-report links. Support hours and plan differences are not stated.The page is undated.Verified · Nov 10, 2026 |
| JetBrains AI | JetBrains AI states AI Free includes unlimited AI completion in JetBrains IDEs and 3 AI credits every 30 days, with no end date stated and no price. The page says it is only available for commercial IDE versions and the FAQ says it is not available in IntelliJ IDEA without an Ultimate subscription, in PyCharm without Pro or in Android Studio. A free 30-day AI Pro trial is offered when you first activate JetBrains AI in a JetBrains IDE.The buy page redirects to https://www.jetbrains.com/store/ai/ and is undated; the footer shows 2026.Provider claim · Nov 10, 2026 | AI Pro: $10.00 per user per month on monthly billing, or $100.00 per user per year on yearly billing (the page says yearly billing saves 2 months); USD, 1 user, with 10 AI credits every 30 days and access to the Junie, Claude Agent, Codex and Gemini agents. The page says taxes may be added and prices shown are approximate.Both figures were read with the page's billing toggle, which only changes the displayed prices. AI Ultimate is $30 per month or $300 per year with 35 credits.Verified · Nov 10, 2026 | A price per user (monthly or yearly) that includes a set number of AI credits, renewed every 30 days from first activation with no roll-over. One AI credit equals USD 1.00 of AI use, and third-party model usage is billed at public API rates with no surcharge. AI Pro and AI Ultimate users can buy top-up credits valid for 12 months. Organizations have no fixed credit plans and buy credits in any amount for shared use.Credit rules are from the page FAQ.Verified · Nov 10, 2026 | JetBrains AI states unlimited AI completion in JetBrains IDEs on AI Free, AI Pro and AI Ultimate. Monthly AI credits (renewed every 30 days): AI Free 3, AI Pro 10, AI Ultimate 35 (including 5 bonus credits). The All Products Pack and dotUltimate subscriptions include 10 credits per 30 days for individuals and 20 credits per seat for organizations.FAQ wording on the All Products Pack is from the same page.Provider claim · Nov 10, 2026 | JetBrains IDEs (completion and agents) and JetBrains Air, which the Air page offers in JetBrains IDEs, on the web, in a CLI and on mobile. The buy page says 25+ third-party agents (for example Cursor, Codex and GitHub Copilot) can be used with your own accounts through the Agent Client Protocol (ACP).The pages read name no plugin for editors other than JetBrains IDEs.Verified · Nov 10, 2026 | The FAQ says JetBrains AI gives access to models from OpenAI, Anthropic, Google Gemini and xAI plus JetBrains' own model for code completion, and agents including Codex, Claude Agent, Gemini CLI and Junie by JetBrains. The page also says you can connect supported providers with an existing provider account or your own API key (Anthropic, OpenAI, Google); that usage is billed by the provider and does not use JetBrains AI credits.Bring-your-own-key wording is from the page FAQ.Verified · Nov 10, 2026 | Access to coding agents in JetBrains Air (Junie, Claude Agent, Codex, Gemini CLI), shown on the AI Pro and AI Ultimate cards; the Air page lists tasks, automations and projects.The Air page is https://www.jetbrains.com/air/.Verified · Nov 10, 2026 | Organizations buy AI credits for shared use and can choose an Air Governance plan: Basic (free; shared credit pool, centralized billing, AI spending limits, SSO and access control, AI analytics), Business ($20.00 per seat per month, adding model and agent controls, group-level policies, central MCP management, an analytics API, managed bring-your-own-key and 24/7 priority support) and Enterprise (custom; adds hybrid deployments, AI audit logs, workspaces, IdP and SCIM configuration). Usage is paid separately, with AI credits or your own provider keys.The Business card shows $20.00 per seat per month with a "Get in touch" button, while the FAQ says "Business (free during early access)". Both are on the same page.Verified · Nov 10, 2026 | JetBrains AI states it does not collect detailed code-related data (the full text of inputs sent to the model and its responses, including source code) from commercial users unless explicit, informed consent has been obtained in advance, that such data can be used for product improvement and for training AI models, and that operational and event data are not used to train its AI. The Basic Air Governance card lists "No training on your data".The notice shows "last updated: July 7, 2026". The Basic card wording is from https://www.jetbrains.com/ai-ides/buy/.Provider claim · Nov 10, 2026 | The Business governance plan lists 24/7 priority support with a first response within 4 hours for critical requests; Enterprise lists first response within 2 hours for critical and 6 hours for standard requests. The page gives no support terms for AI Free, AI Pro or AI Ultimate beyond a "Contact us" form.Response times are as stated on the page.Verified · Nov 10, 2026 |
Questions
How were these AI coding assistants chosen?
The starting list was GitHub Copilot, Cursor, Claude Code, Windsurf, OpenAI Codex, Antigravity, JetBrains AI, Kiro, Amazon Q Developer, Tabnine, Replit, Cline, Zed, Augment Code, Warp, Qodo and Gemini Code Assist. Base rule, applied to every one: the product uses AI to write or edit code for developers and its own pricing page, without a sales call or login, shows a price or a free plan for at least one self-serve plan, and it was sold on the check date. Fifteen met the base rule, which is more than this page holds, so one extra rule was applied to every candidate alike: the product's own pricing page states, in US dollars, both (1) a monthly price for one person on a self-serve paid plan and (2) a monthly price per user or per seat for a plan for teams, with no separate per-team charge added to it. Nine met both and are listed. Set aside by the extra rule: Antigravity (its pricing page shows Google AI Pro and Google AI Ultra without a price), Augment Code (its Standard and Business prices are flat for a whole team, with "no per-seat charge"), Cline (its pricing page shows no prices), Devin, which is the product Windsurf's website now redirects to (its Teams plan adds an $80 per month team charge to the $40 per full seat), Qodo (it prices by credit, not per person or seat) and Replit (its Pro plan is a flat price for up to 15 collaborators). Left out by the base rule: Tabnine, whose pricing address now redirects to a contact form with no plan or price. Gemini Code Assist is not listed because Google's pages say new purchases of its Standard and Enterprise subscriptions ended on October 9, 2026 and its consumer tiers stopped serving requests on June 18, 2026. Claude Code and OpenAI Codex are included because the Claude and ChatGPT pricing pages state prices for plans that include them, and the rows describe the coding product, not the chat plans. This says nothing about the quality of any product left out, and tools that were not on the starting list were not evaluated, so this is not a complete list of AI coding tools. The page does not rank them.
Where do the prices come from?
Each product's own pricing page, plan comparison, docs and privacy pages, in US dollars, read on 2026-11-10. Each price says the plan, the billing term, what usage it includes and any promotion or introductory label. Many of these products charge by usage (credits, dollars of agent usage or token rates), so the monthly price alone does not show what you will pay; the Usage included row gives what each page states. A few pages build prices with scripts, so they were read in a browser, and billing switches and plan tabs were changed only to read other displayed prices. Prices and plans change often, so check the provider before you buy.
How are Claude Code and OpenAI Codex priced here?
They are not sold on their own: Claude Code comes with the paid Claude plans and OpenAI Codex with ChatGPT plans, including the free one for Codex. The rows give the plan prices as the Claude and ChatGPT pricing pages state them and say what each plan includes for coding. The chat features of those plans are not described here.
What does "Not publicly stated" mean?
The provider's pages did not say. It does not mean the answer is no.
Sources
Every fact above links to the page it came from. 28 pages, each with the date we checked it.
- aws.amazon.com/q/developer/ · checked Nov 10, 2026
- aws.amazon.com/q/developer/faqs/ · checked Nov 10, 2026
- aws.amazon.com/q/developer/pricing/ · checked Nov 10, 2026
- chatgpt.com/codex/pricing · checked Nov 10, 2026
- claude.com/pricing · checked Nov 10, 2026
- claude.com/product/claude-code · checked Nov 10, 2026
- code.claude.com/docs/en/data-usage · checked Nov 10, 2026
- code.claude.com/docs/en/overview · checked Nov 10, 2026
- cursor.com/data-use · checked Nov 10, 2026
- cursor.com/docs/models-and-pricing · checked Nov 10, 2026
- cursor.com/download · checked Nov 10, 2026
- cursor.com/pricing · checked Nov 10, 2026
- docs.warp.dev/support-and-community/privacy-and-security/privacy · checked Nov 10, 2026
- github.com/features/copilot/plans · checked Nov 10, 2026
- kiro.dev/docs/privacy-and-security/data-protection/ · checked Nov 10, 2026
- kiro.dev/downloads/ · checked Nov 10, 2026
- kiro.dev/pricing/ · checked Nov 10, 2026
- learn.chatgpt.com/docs/codex/ide · checked Nov 10, 2026
- jetbrains.com/ai-ides/buy/ · checked Nov 10, 2026
- jetbrains.com/air/ · checked Nov 10, 2026
- jetbrains.com/legal/docs/terms/product_data_collection/ · checked Nov 10, 2026
- warp.dev/download · checked Nov 10, 2026
- warp.dev/pricing · checked Nov 10, 2026
- zed.dev/docs/account/zed-hosted-models · checked Nov 10, 2026
- zed.dev/docs/ai/overview · checked Nov 10, 2026
- zed.dev/docs/ai/privacy-and-security · checked Nov 10, 2026
- zed.dev/docs/installation · checked Nov 10, 2026
- zed.dev/pricing · checked Nov 10, 2026
Cite this page: All Things Compared, "AI coding assistants, compared", facts checked Nov 10, 2026. https://allthingscompared.com/ai-coding-assistants/