# AI coding plan calculator: which plan fits

Canonical: https://planyield.dev/calculator/ · Licence CC BY 4.0 · Filter AI coding plans by budget, tools, usage level and weekly caps, on published prices, multiples and credit counts, never hours. Updated 18 Nov 2026.

By [Joe Tay](https://planyield.dev/about/#joe) · Checked 13–18 Nov 2026

Data as of 18 Nov 2026 · Rows checked on vendor pages 13 – 18 Nov 2026 · Page updated 18 Nov 2026

> **Short answer (as of 18 Nov 2026):** For one person, the cheapest AI coding plan at entry level is Alibaba Token Plan Lite at $6 a month. Heavier use depends on the tool, because each multiple is against that product’s own entry plan; the cheapest plan per tool is $43 to $100 a month at about 5x and $199.99 to $200 at about 20x or more.

At about 5x, by tool: Cursor Super at $100, Claude Max at $100, ChatGPT Pro 100 at $100, GitHub Copilot Max at $100, Google AI Ultra $100 (Antigravity) at $99.99, Kiro Pro Max at $100, GLM Coding Plan Pro at $80, and Alibaba Token Plan Pro at $43. At about 20x or more, by tool: Cursor Ultra at $200, ChatGPT Pro 200 at $200, and Google AI Ultra $200 (Antigravity) at $199.99. Choose a budget, tools, usage level and weekly-cap rule below to filter all 52 options. Vendors do not publish limits in hours, so the calculator never converts hours of use into a plan.

**Cite this:** PlanYield, 'AI coding plan calculator: which plan fits', checked 13–18 Nov 2026, https://planyield.dev/calculator/. CC BY 4.0.

## Matching plans

Every plan for one person is in the table below, by price. Choose your needs above and press Show plans to filter it.

## All plans for one person

*42 options, by monthly price, as of 18 Nov 2026. A row with several prices gives one option per price.*

| Product | Plan | Price | Published allowance | Usage multiple | Limit windows | Latest change | Source · checked |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Codex | [ChatGPT Free](https://planyield.dev/plans/codex-chatgpt-free/) | $0 a month | GPT-6 Luna, standard speed, desktop app, subject to rollout; One-off activation (12 Nov 2026 UTC): Tibo says the banked reset has been activated and should land in everyone's accounts sometime soon. INFERRED: the post does not name plans. | Multiple not published | 5-hour: Not listed Weekly: Not listed | 12 Nov [Codex: banked reset activated, rolling out to accounts](https://planyield.dev/changes/2026-11-12-codex-banked-reset-activated/) Vendor post on X | Partly verified [openai.com ↗](https://developers.openai.com/codex/pricing) checked 15 Nov |
| Cursor | [Cursor Free](https://planyield.dev/plans/cursor-free/) | $0 a month | Limited agent + chat requests | Multiple not published | 5-hour: Not listed Weekly: Yes | None logged | Verified [cursor.com ↗](https://cursor.com/pricing) checked 13 Nov |
| Devin | [Devin Free](https://planyield.dev/plans/devin-free/) | $0 a month | Light daily + weekly quota (wait for reset; no extra usage); SWE-2 Flash Medium in Desktop and CLI (High and Max need a paid plan); Devin Review and DeepWiki included | Multiple not published | 5-hour: Not listed Daily: Yes Weekly: Yes | 13 Nov [Devin docs: SWE-2 Flash free until 4 Dec, list price published, Medium on the Free plan](https://planyield.dev/changes/2026-11-13-devin-swe-2-flash-free-to-dec-4/) Vendor page | Verified [devin.ai ↗](https://docs.devin.ai/admin/billing/self-serve) checked 13 Nov |
| Copilot | [GitHub Copilot Free](https://planyield.dev/plans/copilot-free/) | $0 a month | 2,000 completions/mo; limited chat and agent use from an AI-credit allowance (size not published); auto model selection only | Multiple not published | 5-hour: Not listed Weekly: Not listed Monthly: Yes | 19 Oct [Copilot retires GPT-5.5, GPT-5.4 and GPT-5 mini](https://planyield.dev/changes/2026-10-19-copilot-models-retired-oct19/) Vendor page | Verified [github.com ↗](https://github.com/features/copilot/plans) checked 16 Nov |
| Antigravity | [Google Antigravity Free](https://planyield.dev/plans/antigravity-free/) | $0 a month | Basic weekly rate limits; unlimited Tab completions and Command requests; models: Gemini 4 Hydrogen, Gemini 3.8 Flash, Claude Sonnet 4.6 and Opus 4.6 (thinking), GPT-OSS-120b (Claude 4.6 and GPT-OSS-120b until 18 Nov 2026); no Gemini 4 Argon or Gemini 4 Carbon | Multiple not published | 5-hour: Not listed Weekly: Yes | 12 Nov [Antigravity adds Gemini 4 Hydrogen and Orchestrator](https://planyield.dev/changes/2026-11-12-antigravity-gemini-4-hydrogen-orchestrator/) Vendor page | Verified [antigravity.google ↗](https://antigravity.google/pricing) checked 13 Nov |
| Kiro | [Kiro Free](https://planyield.dev/plans/kiro-free/) | $0 a month | 50 credits/mo; Auto (Sonnet-class or better), Claude Sonnet 4.5, and open-weight models (DeepSeek 3.2, MiniMax M2.5, GLM-5, MiniMax M2.1, Qwen3 Coder Next) | Multiple not published | 5-hour: Not listed Weekly: Yes Monthly: Yes | 8 May [Kiro swaps 500-credit sign-up bonus for $20 first-upgrade credit](https://planyield.dev/changes/2026-05-08-kiro-20-signup-bonus/) Vendor announcement | Verified [kiro.dev ↗](https://kiro.dev/pricing/) checked 13 Nov |
| Alibaba Model Studio | [Alibaba Token Plan Lite](https://planyield.dev/plans/alibaba-token-plan-lite/) | $6 a month | 20,000 Credits a month; Qwen (qwen3.8-max, qwen3.8-flash, qwen3.7-max, qwen3.7-plus, qwen3.6-flash, auto), DeepSeek V4/V4.1, GLM-5.3/5.2, image, video and speech models, plus built-in tools; 1–2 concurrent agents | 1x (the baseline) The 1x baseline. The lowest-priced Alibaba Token Plan; the higher tiers are compared by their monthly Credits. From “20,000 Credits a month”. | 5-hour: Not listed Weekly: None (since 22 Sep 2026) Monthly: Yes | 9 Nov [Alibaba raises Token Plan quotas and cuts the Pro price](https://planyield.dev/changes/2026-11-09-alibaba-token-plan-quota-up-pro-cut/) Vendor page | Verified [alibabacloud.com ↗](https://www.alibabacloud.com/help/en/model-studio/token-plan-personal-overview) checked 13 Nov |
| Codex | [ChatGPT Go](https://planyield.dev/plans/codex-chatgpt-go/) | $8 a month | GPT-6 Luna, lightweight tasks, standard speed, desktop app, subject to rollout; One-off activation (12 Nov 2026 UTC): Tibo says the banked reset has been activated and should land in everyone's accounts sometime soon. INFERRED: the post does not name plans. | Multiple not published | 5-hour: Not listed Weekly: Not listed | 12 Nov [Codex: banked reset activated, rolling out to accounts](https://planyield.dev/changes/2026-11-12-codex-banked-reset-activated/) Vendor post on X | Partly verified [openai.com ↗](https://developers.openai.com/codex/pricing) checked 15 Nov |
| Cursor | [Cursor Plus](https://planyield.dev/plans/cursor-plus/) | $8 a month | Basic usage pool (1x); Cursor's own models only (Grok 4.7/4.6/4.5, Composer 2.5) at fixed settings (no Fast mode, medium effort); no third-party models, Auto, Cloud Agents or own API keys | 1x (the baseline) The 1x baseline. From “Basic usage pool (1x)”. | 5-hour: Not listed Weekly: Yes | 13 Oct [Cursor plans move to an X Subscription lineup](https://planyield.dev/changes/2026-10-13-cursor-x-subscription/) Vendor page | Verified [cursor.com ↗](https://cursor.com/pricing) checked 13 Nov |
| Alibaba Model Studio | [Alibaba Token Plan Essential](https://planyield.dev/plans/alibaba-token-plan-essential/) | $10 a month | 45,000 Credits a month; same models and tools as Lite; 2–3 concurrent agents | 2.2x Alibaba Token Plan Lite PlanYield arithmetic on published figures. PlanYield arithmetic: 45,000 ÷ 20,000 Credits a month. From “45,000 Credits a month”; “20,000 Credits a month” (Alibaba Token Plan Lite row). | 5-hour: Not listed Weekly: None (since 22 Sep 2026) Monthly: Yes | 9 Nov [Alibaba raises Token Plan quotas and cuts the Pro price](https://planyield.dev/changes/2026-11-09-alibaba-token-plan-quota-up-pro-cut/) Vendor page | Verified [alibabacloud.com ↗](https://www.alibabacloud.com/help/en/model-studio/token-plan-personal-overview) checked 13 Nov |
| Copilot | [GitHub Copilot Pro](https://planyield.dev/plans/copilot-pro/) | $10 a month | 1,500 AI credits/mo (1,000 base + 500 flex); completions unlimited | 1x (the baseline) The 1x baseline. The lowest-priced paid Copilot plan; the higher tiers are compared by their AI credit counts. From “1,500 AI credits/mo”. | 5-hour: Not listed Weekly: Not listed Monthly: Yes | 27 Oct [GPT-6.1 Terra arrives in Copilot at $1 / $5 per 1M tokens](https://planyield.dev/changes/2026-10-27-copilot-gpt-61-terra/) Vendor page | Verified [github.com ↗](https://docs.github.com/en/copilot/concepts/billing-and-usage/individuals/billing) checked 13 Nov |
| Alibaba Model Studio | [Alibaba Token Plan Standard](https://planyield.dev/plans/alibaba-token-plan-standard/) | $18 a month | 80,000 Credits a month; same models and tools as Lite; 3–4 concurrent agents | 4x Alibaba Token Plan Lite PlanYield arithmetic on published figures. PlanYield arithmetic: 80,000 ÷ 20,000 Credits a month. From “80,000 Credits a month”; “20,000 Credits a month” (Alibaba Token Plan Lite row). | 5-hour: Not listed Weekly: None (since 22 Sep 2026) Monthly: Yes | 9 Nov [Alibaba raises Token Plan quotas and cuts the Pro price](https://planyield.dev/changes/2026-11-09-alibaba-token-plan-quota-up-pro-cut/) Vendor page | Verified [alibabacloud.com ↗](https://www.alibabacloud.com/help/en/model-studio/token-plan-personal-overview) checked 13 Nov |
| GLM Coding Plan | [GLM Coding Plan Lite](https://planyield.dev/plans/glm-coding-plan-lite/) | $18 a month | 2,000 credits per 5 hours and 10,000 a week (the z.ai/subscribe card: '10,000 Credits / week'); built for lightweight iteration on small repos; GLM-5.3 and GLM-5.3-Flash (GLM-5.2/5.1 requests go to GLM-5.3, GLM-4.7 to GLM-5.3-Flash); vision, Web Search, Web Reader and Zread MCP; Z.ai estimates 48–97M GLM-5.3 tokens a week at a 95% cache hit rate | 1x (the baseline) The 1x baseline. The lowest-priced GLM Coding Plan; Pro and Max are published as multiples of Lite usage. From “2,000 credits per 5 hours and 10,000 a week”. | 5-hour: Yes Weekly: Yes | 16 Oct [GLM Coding Plan gets its own API keys](https://planyield.dev/changes/2026-10-16-glm-coding-plan-separate-keys/) Vendor page | Verified [z.ai ↗](https://z.ai/subscribe) checked 15 Nov |
| Kimi Code | [Kimi Code Plus](https://planyield.dev/plans/kimi-code-plus/) | $19 a month | Kimi Code in Desktop, CLI and VS Code, plus an API key for Claude Code, OpenCode, Codex and Hermes Agent; K3.1 Pro, K3 and K2.8 Preview; no 1M context on K3.1 Pro or K3 and no Highspeed models (Pro and above); quota sizes not published | Multiple not published | 5-hour: Yes, rolling Weekly: None (on the new plans) Monthly: Yes | 23 Oct [K3.1 Pro and K3.1 Pro Highspeed arrive in Kimi Code](https://planyield.dev/changes/2026-10-23-kimi-k31-pro/) Vendor page | Partly verified [kimi.com ↗](https://www.kimi.com/membership/pricing) checked 13 Nov |
| Antigravity | [Google AI Pro (Antigravity)](https://planyield.dev/plans/antigravity-google-ai-pro/) | $19.99 a month | Higher weekly rate limit than Free; Gemini 4 Argon (paid Pro, or trial on eligible premium devices); Gemini 4 Hydrogen and Orchestrator (Argon), where Argon plans and Hydrogen implements (from 12 Nov 2026; Orchestrator on paid Pro, or trial on eligible premium devices); no Gemini 4 Carbon; Claude Sonnet 5.5 / Opus 5.5 (thinking) on non-trial Pro only; Gemini 3.8 Flash; Claude 4.6 models and GPT-OSS-120b until 18 Nov 2026; sizes not published | 1x (the baseline) The 1x baseline. From “5x Google AI Pro's rate limit in tokens” (Google AI Ultra $100 (Antigravity) row). | 5-hour: Yes Weekly: Yes | 12 Nov [Antigravity adds Gemini 4 Hydrogen and Orchestrator](https://planyield.dev/changes/2026-11-12-antigravity-gemini-4-hydrogen-orchestrator/) Vendor page | Verified [gemini.google ↗](https://gemini.google/subscriptions/) checked 13 Nov |
| Codex | [ChatGPT Plus](https://planyield.dev/plans/codex-chatgpt-plus/) | $20 a month | Local msgs per 5h (estimates): GPT-6.2 Astra 6-50, GPT-6.2 Sol 17-180, GPT-6.1 Sol 15-160, GPT-6.1 Terra 35-285, GPT-6 Sol 15-150, GPT-6 Astra 5-45, GPT-6 Luna 350-3,000; weekly limits may apply; One-off activation (12 Nov 2026 UTC): Tibo says the banked reset has been activated and should land in everyone's accounts sometime soon. INFERRED: the post does not name plans. | 1x (the baseline) The 1x baseline. From “Pro 500 50x Plus” (ChatGPT Pro row). | 5-hour: Yes Weekly: May apply | 17 Nov [GPT-6.2 Astra and Sol in Codex; Ultrafast on Pro 200](https://planyield.dev/changes/2026-11-17-codex-gpt-62-astra-sol/) Vendor page | Partly verified [openai.com ↗](https://openai.com/index/introducing-gpt-6-2-astra-sol/) checked 18 Nov |
| Claude Code | [Claude Pro](https://planyield.dev/plans/claude-pro/) | $20 a month Listed as $20 monthly; $200/yr up front (card shows $17/mo) | >=5x Free per 5h session; weekly limits on top; no fixed message count; Fable models not in plan limits (usage credits only); Fable 5.2 launch terms (9 Nov 2026 UTC): Pro and Team Standard get a week to try Fable 5.2, up to half your weekly limit, plus one reset. One-off bonus (10 Nov 2026 UTC): all Pro, Max, Ultra and Team users get one free full reset, across all models and limits. Use it anytime before December 10 at 2:00 PM PT. | 1x (the baseline) The 1x baseline. From “5x Pro per 5h session and per week” (Claude Max row). | 5-hour: Yes, rolling Weekly: Yes | 10 Nov [Claude: one free full reset for Pro, Max, Ultra and Team, use before 10 Dec 2:00 PM PT](https://planyield.dev/changes/2026-11-10-claude-free-full-reset/) Vendor post on X | Partly verified [claude.com ↗](https://claude.com/pricing) checked 13 Nov |
| Devin | [Devin Pro](https://planyield.dev/plans/devin-pro/) | $20 a month | Daily + weekly quota (size unpublished); covers Devin Cloud, CLI, Desktop; GPT usage can bill to a linked ChatGPT Go/Plus/Pro plan instead; SWE-2 Flash is free in Desktop and CLI until 4 Dec 2026 18:00 UTC; Cognition says ~1/5 SWE-2's price and ~3x its speed; Devin's docs list SWE-2 Flash at $0.10 input / $0.50 output per 1M tokens after the free period; Blog update (12 Nov 2026): free SWE-2 in Devin Desktop and CLI is being made a permanent part of Teams, Pro and Max. | 1x (the baseline) The 1x baseline. The lowest-priced paid Devin plan; Cognition publishes no multiple for Devin Max. From “Full seat = Pro quota” (Devin Teams row). | 5-hour: Not listed Daily: Yes Weekly: Yes | 13 Nov [Devin docs: SWE-2 Flash free until 4 Dec, list price published, Medium on the Free plan](https://planyield.dev/changes/2026-11-13-devin-swe-2-flash-free-to-dec-4/) Vendor page | Partly verified [devin.ai ↗](https://docs.devin.ai/admin/billing/self-serve) checked 13 Nov |
| Kiro | [Kiro Pro](https://planyield.dev/plans/kiro-pro/) | $20 a month | 1,000 credits/mo; premium models incl. Auto (Opus-class or better), Claude Opus 5.5 (2.0x credits), Sonnet 5.5 (1.3x), GPT-5.6 Sol / Terra / Luna (4.4x / 2.2x / 0.6x) | 1x (the baseline) The 1x baseline. The lowest-priced paid Kiro plan; the higher tiers are compared by their monthly credits. From “1,000 credits/mo”. | 5-hour: Not listed Weekly: Not listed Monthly: Yes | 4 Nov [Kiro adds credit pooling for teams: $20 a seat, 500 shared credits each](https://planyield.dev/changes/2026-11-04-kiro-team-credit-pooling/) Vendor announcement | Verified [kiro.dev ↗](https://kiro.dev/pricing/) checked 13 Nov |
| Cursor | [Cursor Premium](https://planyield.dev/plans/cursor-premium/) | $30 a month | 4x Plus pool; third-party models NOT included in pool | 4x Cursor Plus Published by the vendor. From “4x Plus pool”. | 5-hour: Not listed Weekly: Yes | 13 Oct [Cursor plans move to an X Subscription lineup](https://planyield.dev/changes/2026-10-13-cursor-x-subscription/) Vendor page | Verified [cursor.com ↗](https://cursor.com/pricing) checked 13 Nov |
| Copilot | [GitHub Copilot Pro+](https://planyield.dev/plans/copilot-pro-plus/) | $39 a month | 7,000 AI credits/mo (3,900 base + 3,100 flex); completions unlimited | 4.7x GitHub Copilot Pro PlanYield arithmetic on published figures. PlanYield arithmetic: 7,000 ÷ 1,500 AI credits a month. From “7,000 AI credits/mo”; “1,500 AI credits/mo” (GitHub Copilot Pro row). | 5-hour: Not listed Weekly: Not listed Monthly: Yes | 27 Oct [GPT-6.1 Terra arrives in Copilot at $1 / $5 per 1M tokens](https://planyield.dev/changes/2026-10-27-copilot-gpt-61-terra/) Vendor page | Verified [github.com ↗](https://docs.github.com/en/copilot/concepts/billing-and-usage/individuals/billing) checked 13 Nov |
| Kimi Code | [Kimi Code Pro](https://planyield.dev/plans/kimi-code-pro/) | $39 a month | Everything in Plus, with 1M context on K3.1 Pro and K3 and the Highspeed models (K3.1 Pro Highspeed, K2.7 Code Highspeed); quota sizes not published | Multiple not published | 5-hour: Yes, rolling Weekly: None (on the new plans) Monthly: Yes | 23 Oct [K3.1 Pro and K3.1 Pro Highspeed arrive in Kimi Code](https://planyield.dev/changes/2026-10-23-kimi-k31-pro/) Vendor page | Partly verified [kimi.com ↗](https://www.kimi.com/membership/pricing) checked 13 Nov |
| Kiro | [Kiro Pro+](https://planyield.dev/plans/kiro-pro-plus/) | $40 a month | 2,000 credits/mo; same models as Pro | 2x Kiro Pro PlanYield arithmetic on published figures. PlanYield arithmetic: 2,000 ÷ 1,000 credits a month. From “2,000 credits/mo”; “1,000 credits/mo” (Kiro Pro row). | 5-hour: Not listed Weekly: Not listed Monthly: Yes | 4 Nov [Kiro adds credit pooling for teams: $20 a seat, 500 shared credits each](https://planyield.dev/changes/2026-11-04-kiro-team-credit-pooling/) Vendor announcement | Verified [kiro.dev ↗](https://kiro.dev/pricing/) checked 13 Nov |
| Alibaba Model Studio | [Alibaba Token Plan Pro](https://planyield.dev/plans/alibaba-token-plan-pro/) | $43 a month | 200,000 Credits a month; same models and tools as Lite; 6–8 concurrent agents | 10x Alibaba Token Plan Lite PlanYield arithmetic on published figures. PlanYield arithmetic: 200,000 ÷ 20,000 Credits a month. From “200,000 Credits a month”; “20,000 Credits a month” (Alibaba Token Plan Lite row). | 5-hour: Not listed Weekly: None (since 22 Sep 2026) Monthly: Yes | 9 Nov [Alibaba raises Token Plan quotas and cuts the Pro price](https://planyield.dev/changes/2026-11-09-alibaba-token-plan-quota-up-pro-cut/) Vendor page | Verified [alibabacloud.com ↗](https://www.alibabacloud.com/help/en/model-studio/token-plan-personal-overview) checked 13 Nov |
| Alibaba Model Studio | [Alibaba Coding Plan Pro](https://planyield.dev/plans/alibaba-coding-plan-pro/) | $50 a month | 6,000 requests per 5 hours, 45,000 a week and 90,000 a month (one request is one model call; Alibaba says simple tasks use 5–10 calls, complex ones 10–30 or more); models: qwen3.7-plus, qwen3.6-plus, kimi-k2.5, glm-5, MiniMax-M2.5 and qwen3.5-plus only | Multiple not published | 5-hour: Yes Weekly: Yes Monthly: Yes | None logged | Verified [alibabacloud.com ↗](https://www.alibabacloud.com/help/en/model-studio/coding-plan) checked 13 Nov |
| GLM Coding Plan | [GLM Coding Plan Pro](https://planyield.dev/plans/glm-coding-plan-pro/) | $80 a month | 12,000 credits per 5 hours and 60,000 a week (the z.ai/subscribe card: '6× Lite usage'); all Lite benefits; built for day-to-day development on mid-sized repos; GLM-5.3 and GLM-5.3-Flash (GLM-5.2/5.1 requests go to GLM-5.3, GLM-4.7 to GLM-5.3-Flash); vision, Web Search, Web Reader and Zread MCP; a curated selection of MCP tools; faster generation speeds; priority access to the latest flagship models and features; Z.ai estimates 290–580M GLM-5.3 tokens a week at a 95% cache hit rate | 6x GLM Coding Plan Lite Published by the vendor. From “6× Lite usage”. | 5-hour: Yes Weekly: Yes | 16 Oct [GLM Coding Plan gets its own API keys](https://planyield.dev/changes/2026-10-16-glm-coding-plan-separate-keys/) Vendor page | Verified [z.ai ↗](https://z.ai/subscribe) checked 15 Nov |
| Kimi Code | [Kimi Code Max](https://planyield.dev/plans/kimi-code-max/) | $99 a month | Same models and context as Pro; a larger quota than Pro, size not published | Multiple not published | 5-hour: Yes, rolling Weekly: None (on the new plans) Monthly: Yes | 23 Oct [K3.1 Pro and K3.1 Pro Highspeed arrive in Kimi Code](https://planyield.dev/changes/2026-10-23-kimi-k31-pro/) Vendor page | Partly verified [kimi.com ↗](https://www.kimi.com/membership/pricing) checked 13 Nov |
| Antigravity | [Google AI Ultra $100 (Antigravity)](https://planyield.dev/plans/antigravity-google-ai-ultra-5x/) | $99.99 a month | 5x Google AI Pro's rate limit in tokens; priority access; Gemini 4 Carbon (Ultra only, from 10 Nov 2026), Gemini 4 Argon, Gemini 4 Hydrogen, Orchestrator (Carbon) and Orchestrator (Argon) (from 12 Nov 2026), Gemini 3.8 Flash, Claude Sonnet 5.5 / Opus 5.5 (thinking), GPT-OSS-120b (until 18 Nov 2026) | 5x Google AI Pro (Antigravity) Published by the vendor. From “5x Google AI Pro's rate limit in tokens”. | 5-hour: Yes Weekly: Yes | 12 Nov [Antigravity adds Gemini 4 Hydrogen and Orchestrator](https://planyield.dev/changes/2026-11-12-antigravity-gemini-4-hydrogen-orchestrator/) Vendor page | Verified [gemini.google ↗](https://gemini.google/subscriptions/) checked 13 Nov |
| Codex | [ChatGPT Pro 100](https://planyield.dev/plans/codex-chatgpt-pro/) | $100 a month Listed as $100 / $200 / $500 | Pro 100: 10x Plus (doubled on 29 Oct US) Earlier (reported): 5x Plus. Weekly size not published. | 10x ChatGPT Plus Reported, not stated on the vendor's page. 10x Plus confirmed in the 8 Nov staff post; absolute weekly size unpublished. From “Pro 100 10x Plus”. | 5-hour: None Weekly: Yes | 17 Nov [GPT-6.2 Astra and Sol in Codex; Ultrafast on Pro 200](https://planyield.dev/changes/2026-11-17-codex-gpt-62-astra-sol/) Vendor page | Partly verified [openai.com ↗](https://openai.com/index/introducing-gpt-6-2-astra-sol/) checked 18 Nov |
| Claude Code | [Claude Max](https://planyield.dev/plans/claude-max-5x/) | $100 a month | 5x Pro per 5h session and per week; Fable models capped at 50% of weekly limits. One-off bonus (10 Nov 2026 UTC): all Pro, Max, Ultra and Team users get one free full reset, across all models and limits. Use it anytime before December 10 at 2:00 PM PT. | 5x Claude Pro per 5-hour session and per week Published by the vendor. From “5x Pro per 5h session and per week”. | 5-hour: Yes Weekly: Yes | 10 Nov [Claude: one free full reset for Pro, Max, Ultra and Team, use before 10 Dec 2:00 PM PT](https://planyield.dev/changes/2026-11-10-claude-free-full-reset/) Vendor post on X | Partly verified [claude.com ↗](https://claude.com/pricing) checked 13 Nov |
| Cursor | [Cursor Super](https://planyield.dev/plans/cursor-super/) | $100 a month | 15x Plus; priority at peak | 15x Cursor Plus Published by the vendor. From “15x Plus”. | 5-hour: Not listed Weekly: Yes | 13 Oct [Cursor plans move to an X Subscription lineup](https://planyield.dev/changes/2026-10-13-cursor-x-subscription/) Vendor page | Verified [cursor.com ↗](https://cursor.com/pricing) checked 13 Nov |
| Copilot | [GitHub Copilot Max](https://planyield.dev/plans/copilot-max/) | $100 a month | 20,000 AI credits/mo (10,000 base + 10,000 flex); completions unlimited | 13.3x GitHub Copilot Pro PlanYield arithmetic on published figures. PlanYield arithmetic: 20,000 ÷ 1,500 AI credits a month. From “20,000 AI credits/mo”; “1,500 AI credits/mo” (GitHub Copilot Pro row). | 5-hour: Not listed Weekly: Not listed Monthly: Yes | 29 Oct [Copilot Max's Auto discount returns to 10%](https://planyield.dev/changes/2026-10-29-copilot-max-auto-discount-10/) Vendor page | Verified [github.com ↗](https://docs.github.com/en/copilot/concepts/billing-and-usage/individuals/billing) checked 13 Nov |
| Kiro | [Kiro Pro Max](https://planyield.dev/plans/kiro-pro-max/) | $100 a month | 5,000 credits/mo (2.5x Pro+); same models as Pro | 5x Kiro Pro PlanYield arithmetic on published figures. PlanYield arithmetic: 5,000 ÷ 1,000 credits a month. From “5,000 credits/mo”; “1,000 credits/mo” (Kiro Pro row). | 5-hour: Not listed Weekly: Not listed Monthly: Yes | 4 Nov [Kiro adds credit pooling for teams: $20 a seat, 500 shared credits each](https://planyield.dev/changes/2026-11-04-kiro-team-credit-pooling/) Vendor announcement | Verified [kiro.dev ↗](https://kiro.dev/pricing/) checked 13 Nov |
| GLM Coding Plan | [GLM Coding Plan Max](https://planyield.dev/plans/glm-coding-plan-max/) | $168 a month | 28,000 credits per 5 hours and 140,000 a week (the z.ai/subscribe card: '14× Lite usage'); all Pro benefits; built for advanced users working on mid-to-large repos; GLM-5.3 and GLM-5.3-Flash (GLM-5.2/5.1 requests go to GLM-5.3, GLM-4.7 to GLM-5.3-Flash); vision, Web Search, Web Reader and Zread MCP; first access to the latest flagship models and features; dedicated resources during peak times; Z.ai estimates 676–1,352M GLM-5.3 tokens a week at a 95% cache hit rate | 14x GLM Coding Plan Lite Published by the vendor. From “14× Lite usage”. | 5-hour: Yes Weekly: Yes | 16 Oct [GLM Coding Plan gets its own API keys](https://planyield.dev/changes/2026-10-16-glm-coding-plan-separate-keys/) Vendor page | Verified [z.ai ↗](https://z.ai/subscribe) checked 15 Nov |
| Kimi Code | [Kimi Code Ultra](https://planyield.dev/plans/kimi-code-ultra/) | $199 a month | Same models and context as Pro; the largest quota, size not published | Multiple not published | 5-hour: Yes, rolling Weekly: None (on the new plans) Monthly: Yes | 23 Oct [K3.1 Pro and K3.1 Pro Highspeed arrive in Kimi Code](https://planyield.dev/changes/2026-10-23-kimi-k31-pro/) Vendor page | Partly verified [kimi.com ↗](https://www.kimi.com/membership/pricing) checked 13 Nov |
| Antigravity | [Google AI Ultra $200 (Antigravity)](https://planyield.dev/plans/antigravity-google-ai-ultra-20x/) | $199.99 a month | 20x Google AI Pro's rate limit in tokens; highest rate limits; same models as Ultra 5x | 20x Google AI Pro (Antigravity) Published by the vendor. From “20x Google AI Pro's rate limit in tokens”. | 5-hour: Yes Weekly: Yes | 12 Nov [Antigravity adds Gemini 4 Hydrogen and Orchestrator](https://planyield.dev/changes/2026-11-12-antigravity-gemini-4-hydrogen-orchestrator/) Vendor page | Verified [gemini.google ↗](https://gemini.google/subscriptions/) checked 13 Nov |
| Codex | [ChatGPT Pro 200](https://planyield.dev/plans/codex-chatgpt-pro/) | $200 a month Listed as $100 / $200 / $500 | Pro 200: 20x Plus (cut cancelled, every Pro 200 subscriber now on the higher allowance) Earlier (reported): 10x Plus (20x for some to 29 Oct). Weekly size not published. | 20x ChatGPT Plus Reported, not stated on the vendor's page. OpenAI cancelled the Pro 200 cut, and its Help Center says subscribers on the lower allowance had it doubled to match the higher one, so every Pro 200 subscriber now has the higher allowance, confirmed as 20x Plus in the 8 Nov staff post; absolute weekly size unpublished. From “Pro 200 20x Plus”. | 5-hour: None Weekly: Yes | 17 Nov [GPT-6.2 Astra and Sol in Codex; Ultrafast on Pro 200](https://planyield.dev/changes/2026-11-17-codex-gpt-62-astra-sol/) Vendor page | Partly verified [openai.com ↗](https://openai.com/index/introducing-gpt-6-2-astra-sol/) checked 18 Nov |
| Claude Code | [Claude Ultra](https://planyield.dev/plans/claude-max-20x/) | $200 a month | 10x Pro per week; no 5-hour limit for verified accounts (newer accounts keep a temporary one until verified); Fable models capped at 50% of weekly limits; Fable 5.2 launch terms (9 Nov 2026 UTC): Ultra and Team Premium users get Fable resets based on how long you've subscribed: <6 mo: 1; 6–12 mo: 2; >12 mo: 3. Use them anytime by March 8. One-off bonus (10 Nov 2026 UTC): all Pro, Max, Ultra and Team users get one free full reset, across all models and limits. Use it anytime before December 10 at 2:00 PM PT. | 10x Claude Pro per week Published by the vendor. Weekly only: Ultra has no 5-hour limit for verified accounts. From “10x Pro per week”. | 5-hour: None (for verified accounts) Weekly: Yes | 10 Nov [Claude: one free full reset for Pro, Max, Ultra and Team, use before 10 Dec 2:00 PM PT](https://planyield.dev/changes/2026-11-10-claude-free-full-reset/) Vendor post on X | Partly verified [claude.com ↗](https://claude.com/pricing) checked 13 Nov |
| Cursor | [Cursor Ultra](https://planyield.dev/plans/cursor-ultra/) | $200 a month | 50x Plus | 50x Cursor Plus Published by the vendor. From “50x Plus”. | 5-hour: Not listed Weekly: Yes | 13 Oct [Cursor plans move to an X Subscription lineup](https://planyield.dev/changes/2026-10-13-cursor-x-subscription/) Vendor page | Verified [cursor.com ↗](https://cursor.com/pricing) checked 13 Nov |
| Devin | [Devin Max](https://planyield.dev/plans/devin-max/) | $200 a month | Significantly larger weekly quota, no daily cap; GPT usage can bill to a linked ChatGPT Go/Plus/Pro plan instead; SWE-2 Flash is free in Desktop and CLI until 4 Dec 2026 18:00 UTC; Cognition says ~1/5 SWE-2's price and ~3x its speed; Devin's docs list SWE-2 Flash at $0.10 input / $0.50 output per 1M tokens after the free period; Blog update (12 Nov 2026): free SWE-2 in Devin Desktop and CLI is being made a permanent part of Teams, Pro and Max. | Multiple not published | 5-hour: Not listed Daily: None (no daily cap) Weekly: Yes | 13 Nov [Devin docs: SWE-2 Flash free until 4 Dec, list price published, Medium on the Free plan](https://planyield.dev/changes/2026-11-13-devin-swe-2-flash-free-to-dec-4/) Vendor page | Partly verified [devin.ai ↗](https://docs.devin.ai/admin/billing/self-serve) checked 13 Nov |
| Kiro | [Kiro Power](https://planyield.dev/plans/kiro-power/) | $200 a month | 10,000 credits/mo; same models as Pro | 10x Kiro Pro PlanYield arithmetic on published figures. PlanYield arithmetic: 10,000 ÷ 1,000 credits a month. From “10,000 credits/mo”; “1,000 credits/mo” (Kiro Pro row). | 5-hour: Not listed Weekly: Not listed Monthly: Yes | 4 Nov [Kiro adds credit pooling for teams: $20 a seat, 500 shared credits each](https://planyield.dev/changes/2026-11-04-kiro-team-credit-pooling/) Vendor announcement | Verified [kiro.dev ↗](https://kiro.dev/pricing/) checked 13 Nov |
| Codex | [ChatGPT Pro 500](https://planyield.dev/plans/codex-chatgpt-pro/) | $500 a month Listed as $100 / $200 / $500 | Pro 500: 50x Plus since 29 Oct (US) Earlier (reported): 25x Plus. Weekly size not published. | 50x ChatGPT Plus Reported, not stated on the vendor's page. 50x Plus confirmed in the 8 Nov staff post; OpenAI's Help Center states no multiple; absolute weekly size unpublished. From “Pro 500 50x Plus”. | 5-hour: None Weekly: Yes | 17 Nov [GPT-6.2 Astra and Sol in Codex; Ultrafast on Pro 200](https://planyield.dev/changes/2026-11-17-codex-gpt-62-astra-sol/) Vendor page | Partly verified [openai.com ↗](https://openai.com/index/introducing-gpt-6-2-astra-sol/) checked 18 Nov |

## Team and business plans

*Per-seat options, by monthly price, as of 18 Nov 2026. Prices are the monthly figures each row lists.*

| Product | Plan | Price | Published allowance | Usage multiple | Limit windows | Latest change | Source · checked |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Copilot | [GitHub Copilot Business](https://planyield.dev/plans/copilot-business/) | $19 a month Listed as $19 per granted seat/mo | 1,900 AI credits per user a month, pooled across the organization or enterprise (100 users share 190,000); GitHub publishes no base/flex split for Business; inline completions and next edit suggestions unlimited | 1.3x GitHub Copilot Pro PlanYield arithmetic on published figures. PlanYield arithmetic: 1,900 AI credits per user ÷ 1,500 a month. Business pools its credits across the organization, so one person can use more or less than 1,900. From “1,900 AI credits per user a month”; “1,500 AI credits/mo” (GitHub Copilot Pro row). | 5-hour: Not listed Weekly: Not listed Monthly: Yes | None logged | Verified [github.com ↗](https://docs.github.com/en/copilot/get-started/plans) checked 13 Nov |
| Codex | [ChatGPT Business, Standard seat](https://planyield.dev/plans/codex-chatgpt-business/) | $25 a month Listed as Standard $20/user/month billed annually or $25 monthly; Premium $100/user/month billed annually or $125 monthly (2+ seats; seat types can be mixed) | Standard: same 5h estimates as Plus (GPT-6.2 Astra 6-50, GPT-6.2 Sol 17-180 local messages); Premium: 5x Standard usage before 29 Oct 2026 (US), when OpenAI doubled Business Premium limits; no 5-hour limit; One-off activation (12 Nov 2026 UTC): Tibo says the banked reset has been activated and should land in everyone's accounts sometime soon. INFERRED: the post does not name plans. | 1x ChatGPT Plus Published by the vendor. From “Standard: same 5h estimates as Plus”. | 5-hour: Yes Weekly: Yes | 17 Nov [GPT-6.2 Astra and Sol in Codex; Ultrafast on Pro 200](https://planyield.dev/changes/2026-11-17-codex-gpt-62-astra-sol/) Vendor page | Partly verified [openai.com ↗](https://openai.com/index/introducing-gpt-6-2-astra-sol/) checked 18 Nov |
| Claude Code | [Claude Team, Standard seat](https://planyield.dev/plans/claude-team/) | $25 a month Listed as $20 annual / $25 monthly (Standard); $100 annual / $125 monthly (Premium) | Standard > Pro; Premium = 5x Standard; per 5-hour session, Standard = 1.25x Pro and Premium = 6.25x Pro; Standard seats: Fable via usage credits only; Premium seats: Fable up to 50% of weekly limits; Fable 5.2 launch terms (9 Nov 2026 UTC): Ultra and Team Premium users get Fable resets based on how long you've subscribed: <6 mo: 1; 6–12 mo: 2; >12 mo: 3. Use them anytime by March 8. Pro and Team Standard get a week to try Fable 5.2, up to half your weekly limit, plus one reset. One-off bonus (10 Nov 2026 UTC): all Pro, Max, Ultra and Team users get one free full reset, across all models and limits. Use it anytime before December 10 at 2:00 PM PT. | more than 1x Claude Pro Published by the vendor. From “Standard > Pro”. | 5-hour: Yes Weekly: Yes | 10 Nov [Claude: one free full reset for Pro, Max, Ultra and Team, use before 10 Dec 2:00 PM PT](https://planyield.dev/changes/2026-11-10-claude-free-full-reset/) Vendor post on X | Partly verified [claude.com ↗](https://claude.com/pricing) checked 13 Nov |
| Cursor | [Cursor Teams, Premium seat](https://planyield.dev/plans/cursor-teams/) | $30 a month Listed as $30, $100 or $200 per seat/mo | Premium, Super or Ultra seats (4x, 15x or 50x Plus), mixable on one team; Plus not available; free seats have no team allowance | 4x Cursor Plus Published by the vendor. From “Premium, Super or Ultra seats (4x, 15x or 50x Plus)”. | 5-hour: Not listed Weekly: Yes | 13 Oct [Cursor plans move to an X Subscription lineup](https://planyield.dev/changes/2026-10-13-cursor-x-subscription/) Vendor page | Verified [cursor.com ↗](https://cursor.com/docs/account/teams/pricing) checked 13 Nov |
| Copilot | [GitHub Copilot Enterprise](https://planyield.dev/plans/copilot-enterprise/) | $39 a month Listed as $39 per granted seat/mo | 3,900 AI credits per user a month, pooled across the enterprise; GitHub Enterprise Cloud only; priority access to premium and new models; inline completions and next edit suggestions unlimited | 2.6x GitHub Copilot Pro PlanYield arithmetic on published figures. PlanYield arithmetic: 3,900 AI credits per user ÷ 1,500 a month. Enterprise pools its credits across the enterprise, so one person can use more or less than 3,900. From “3,900 AI credits per user a month”; “1,500 AI credits/mo” (GitHub Copilot Pro row). | 5-hour: Not listed Weekly: Not listed Monthly: Yes | None logged | Verified [github.com ↗](https://docs.github.com/en/copilot/get-started/plans) checked 13 Nov |
| Devin | [Devin Teams](https://planyield.dev/plans/devin-teams/) | $80 a month Listed as $80/mo minimum; full seats $40/mo each; flex seats free | Full seat = Pro quota; flex seats draw from shared credits; each member can link their own ChatGPT plan for GPT usage; SWE-2 Flash is free in Desktop and CLI until 4 Dec 2026 18:00 UTC; Cognition says ~1/5 SWE-2's price and ~3x its speed; Devin's docs list SWE-2 Flash at $0.10 input / $0.50 output per 1M tokens after the free period; Blog update (12 Nov 2026): Devin Desktop included for everyone on the team, without paying for extra Desktop seats; new Free mode powered by SWE-2 gives every Teams user 2 concurrent Free sessions, with the option to pay for more; free SWE-2 in Devin Desktop and CLI is being made a permanent part of Teams, Pro and Max; SSO with Okta, Microsoft Entra ID, or any other SAML or OIDC provider, admins can turn off email login once SSO is connected; admins can set a usage limit for each user, with up to 3 tiers so different groups can have different limits; requests for more usage can wait for admin approval or be approved automatically up to an admin-set maximum; standard Business Associate Agreement (BAA) available on Teams without an enterprise contract, requestable by admins from Settings; Teams plans are now limited to 100 seats, down from 200. | 1x Devin Pro Published by the vendor. From “Full seat = Pro quota”. | 5-hour: Not listed Daily: Yes Weekly: Yes | 13 Nov [Devin docs: SWE-2 Flash free until 4 Dec, list price published, Medium on the Free plan](https://planyield.dev/changes/2026-11-13-devin-swe-2-flash-free-to-dec-4/) Vendor page | Partly verified [devin.ai ↗](https://docs.devin.ai/admin/billing/self-serve) checked 13 Nov |
| Cursor | [Cursor Teams, Super seat](https://planyield.dev/plans/cursor-teams/) | $100 a month Listed as $30, $100 or $200 per seat/mo | Premium, Super or Ultra seats (4x, 15x or 50x Plus), mixable on one team; Plus not available; free seats have no team allowance | 15x Cursor Plus Published by the vendor. From “Premium, Super or Ultra seats (4x, 15x or 50x Plus)”. | 5-hour: Not listed Weekly: Yes | 13 Oct [Cursor plans move to an X Subscription lineup](https://planyield.dev/changes/2026-10-13-cursor-x-subscription/) Vendor page | Verified [cursor.com ↗](https://cursor.com/docs/account/teams/pricing) checked 13 Nov |
| Codex | [ChatGPT Business, Premium seat](https://planyield.dev/plans/codex-chatgpt-business/) | $125 a month Listed as Standard $20/user/month billed annually or $25 monthly; Premium $100/user/month billed annually or $125 monthly (2+ seats; seat types can be mixed) | Standard: same 5h estimates as Plus (GPT-6.2 Astra 6-50, GPT-6.2 Sol 17-180 local messages); Premium: 5x Standard usage before 29 Oct 2026 (US), when OpenAI doubled Business Premium limits; no 5-hour limit; One-off activation (12 Nov 2026 UTC): Tibo says the banked reset has been activated and should land in everyone's accounts sometime soon. INFERRED: the post does not name plans. | at least 5x ChatGPT Plus PlanYield arithmetic on published figures. PlanYield arithmetic: 5x Standard before 29 Oct, and Standard matches Plus, so at least 5x Plus. OpenAI staff say Premium limits doubled on 29 Oct; the new size is not stated. From “Premium: 5x Standard usage before 29 Oct 2026”; “Standard: same 5h estimates as Plus”. | 5-hour: None Weekly: Yes | 17 Nov [GPT-6.2 Astra and Sol in Codex; Ultrafast on Pro 200](https://planyield.dev/changes/2026-11-17-codex-gpt-62-astra-sol/) Vendor page | Partly verified [openai.com ↗](https://openai.com/index/introducing-gpt-6-2-astra-sol/) checked 18 Nov |
| Claude Code | [Claude Team, Premium seat](https://planyield.dev/plans/claude-team/) | $125 a month Listed as $20 annual / $25 monthly (Standard); $100 annual / $125 monthly (Premium) | Standard > Pro; Premium = 5x Standard; per 5-hour session, Standard = 1.25x Pro and Premium = 6.25x Pro; Standard seats: Fable via usage credits only; Premium seats: Fable up to 50% of weekly limits; Fable 5.2 launch terms (9 Nov 2026 UTC): Ultra and Team Premium users get Fable resets based on how long you've subscribed: <6 mo: 1; 6–12 mo: 2; >12 mo: 3. Use them anytime by March 8. Pro and Team Standard get a week to try Fable 5.2, up to half your weekly limit, plus one reset. One-off bonus (10 Nov 2026 UTC): all Pro, Max, Ultra and Team users get one free full reset, across all models and limits. Use it anytime before December 10 at 2:00 PM PT. | more than 5x Claude Pro PlanYield arithmetic on published figures. PlanYield arithmetic: Premium is 5x Standard and Standard is more than Pro, so Premium is more than 5x Pro. From “Premium = 5x Standard”; “Standard > Pro”. | 5-hour: Yes Weekly: Yes | 10 Nov [Claude: one free full reset for Pro, Max, Ultra and Team, use before 10 Dec 2:00 PM PT](https://planyield.dev/changes/2026-11-10-claude-free-full-reset/) Vendor post on X | Partly verified [claude.com ↗](https://claude.com/pricing) checked 13 Nov |
| Cursor | [Cursor Teams, Ultra seat](https://planyield.dev/plans/cursor-teams/) | $200 a month Listed as $30, $100 or $200 per seat/mo | Premium, Super or Ultra seats (4x, 15x or 50x Plus), mixable on one team; Plus not available; free seats have no team allowance | 50x Cursor Plus Published by the vendor. From “Premium, Super or Ultra seats (4x, 15x or 50x Plus)”. | 5-hour: Not listed Weekly: Yes | 13 Oct [Cursor plans move to an X Subscription lineup](https://planyield.dev/changes/2026-10-13-cursor-x-subscription/) Vendor page | Verified [cursor.com ↗](https://cursor.com/docs/account/teams/pricing) checked 13 Nov |

## Where each multiple comes from

Every [usage multiple](https://planyield.dev/glossary/#usage-multiple) the calculator uses, with the text it is read from. Each is against that vendor’s own entry paid plan, so a multiple from one vendor does not convert into another’s.

- Cursor Plus: 1x (the baseline). The 1x baseline. From “Basic usage pool (1x)”.
- Cursor Premium: 4x Cursor Plus. Published by the vendor. From “4x Plus pool”.
- Cursor Teams, Premium seat: 4x Cursor Plus. Published by the vendor. From “Premium, Super or Ultra seats (4x, 15x or 50x Plus)”.
- Cursor Super: 15x Cursor Plus. Published by the vendor. From “15x Plus”.
- Cursor Teams, Super seat: 15x Cursor Plus. Published by the vendor. From “Premium, Super or Ultra seats (4x, 15x or 50x Plus)”.
- Cursor Teams, Ultra seat: 50x Cursor Plus. Published by the vendor. From “Premium, Super or Ultra seats (4x, 15x or 50x Plus)”.
- Cursor Ultra: 50x Cursor Plus. Published by the vendor. From “50x Plus”.
- Claude Pro: 1x (the baseline). The 1x baseline. From “5x Pro per 5h session and per week” (Claude Max row).
- Claude Team, Standard seat: more than 1x Claude Pro. Published by the vendor. From “Standard > Pro”.
- Claude Max: 5x Claude Pro per 5-hour session and per week. Published by the vendor. From “5x Pro per 5h session and per week”.
- Claude Team, Premium seat: more than 5x Claude Pro. PlanYield arithmetic on published figures. PlanYield arithmetic: Premium is 5x Standard and Standard is more than Pro, so Premium is more than 5x Pro. From “Premium = 5x Standard”; “Standard > Pro”.
- Claude Ultra: 10x Claude Pro per week. Published by the vendor. Weekly only: Ultra has no 5-hour limit for verified accounts. From “10x Pro per week”.
- ChatGPT Plus: 1x (the baseline). The 1x baseline. From “Pro 500 50x Plus” (ChatGPT Pro row).
- ChatGPT Business, Standard seat: 1x ChatGPT Plus. Published by the vendor. From “Standard: same 5h estimates as Plus”.
- ChatGPT Pro 100: 10x ChatGPT Plus. Reported, not stated on the vendor's page. 10x Plus confirmed in the 8 Nov staff post; absolute weekly size unpublished. From “Pro 100 10x Plus”.
- ChatGPT Business, Premium seat: at least 5x ChatGPT Plus. PlanYield arithmetic on published figures. PlanYield arithmetic: 5x Standard before 29 Oct, and Standard matches Plus, so at least 5x Plus. OpenAI staff say Premium limits doubled on 29 Oct; the new size is not stated. From “Premium: 5x Standard usage before 29 Oct 2026”; “Standard: same 5h estimates as Plus”.
- ChatGPT Pro 200: 20x ChatGPT Plus. Reported, not stated on the vendor's page. OpenAI cancelled the Pro 200 cut, and its Help Center says subscribers on the lower allowance had it doubled to match the higher one, so every Pro 200 subscriber now has the higher allowance, confirmed as 20x Plus in the 8 Nov staff post; absolute weekly size unpublished. From “Pro 200 20x Plus”.
- ChatGPT Pro 500: 50x ChatGPT Plus. Reported, not stated on the vendor's page. 50x Plus confirmed in the 8 Nov staff post; OpenAI's Help Center states no multiple; absolute weekly size unpublished. From “Pro 500 50x Plus”.
- GitHub Copilot Pro: 1x (the baseline). The 1x baseline. The lowest-priced paid Copilot plan; the higher tiers are compared by their AI credit counts. From “1,500 AI credits/mo”.
- GitHub Copilot Business: 1.3x GitHub Copilot Pro. PlanYield arithmetic on published figures. PlanYield arithmetic: 1,900 AI credits per user ÷ 1,500 a month. Business pools its credits across the organization, so one person can use more or less than 1,900. From “1,900 AI credits per user a month”; “1,500 AI credits/mo” (GitHub Copilot Pro row).
- GitHub Copilot Enterprise: 2.6x GitHub Copilot Pro. PlanYield arithmetic on published figures. PlanYield arithmetic: 3,900 AI credits per user ÷ 1,500 a month. Enterprise pools its credits across the enterprise, so one person can use more or less than 3,900. From “3,900 AI credits per user a month”; “1,500 AI credits/mo” (GitHub Copilot Pro row).
- GitHub Copilot Pro+: 4.7x GitHub Copilot Pro. PlanYield arithmetic on published figures. PlanYield arithmetic: 7,000 ÷ 1,500 AI credits a month. From “7,000 AI credits/mo”; “1,500 AI credits/mo” (GitHub Copilot Pro row).
- GitHub Copilot Max: 13.3x GitHub Copilot Pro. PlanYield arithmetic on published figures. PlanYield arithmetic: 20,000 ÷ 1,500 AI credits a month. From “20,000 AI credits/mo”; “1,500 AI credits/mo” (GitHub Copilot Pro row).
- Devin Pro: 1x (the baseline). The 1x baseline. The lowest-priced paid Devin plan; Cognition publishes no multiple for Devin Max. From “Full seat = Pro quota” (Devin Teams row).
- Devin Teams: 1x Devin Pro. Published by the vendor. From “Full seat = Pro quota”.
- Google AI Pro (Antigravity): 1x (the baseline). The 1x baseline. From “5x Google AI Pro's rate limit in tokens” (Google AI Ultra $100 (Antigravity) row).
- Google AI Ultra $100 (Antigravity): 5x Google AI Pro (Antigravity). Published by the vendor. From “5x Google AI Pro's rate limit in tokens”.
- Google AI Ultra $200 (Antigravity): 20x Google AI Pro (Antigravity). Published by the vendor. From “20x Google AI Pro's rate limit in tokens”.
- Kiro Pro: 1x (the baseline). The 1x baseline. The lowest-priced paid Kiro plan; the higher tiers are compared by their monthly credits. From “1,000 credits/mo”.
- Kiro Pro+: 2x Kiro Pro. PlanYield arithmetic on published figures. PlanYield arithmetic: 2,000 ÷ 1,000 credits a month. From “2,000 credits/mo”; “1,000 credits/mo” (Kiro Pro row).
- Kiro Pro Max: 5x Kiro Pro. PlanYield arithmetic on published figures. PlanYield arithmetic: 5,000 ÷ 1,000 credits a month. From “5,000 credits/mo”; “1,000 credits/mo” (Kiro Pro row).
- Kiro Power: 10x Kiro Pro. PlanYield arithmetic on published figures. PlanYield arithmetic: 10,000 ÷ 1,000 credits a month. From “10,000 credits/mo”; “1,000 credits/mo” (Kiro Pro row).
- GLM Coding Plan Lite: 1x (the baseline). The 1x baseline. The lowest-priced GLM Coding Plan; Pro and Max are published as multiples of Lite usage. From “2,000 credits per 5 hours and 10,000 a week”.
- GLM Coding Plan Pro: 6x GLM Coding Plan Lite. Published by the vendor. From “6× Lite usage”.
- GLM Coding Plan Max: 14x GLM Coding Plan Lite. Published by the vendor. From “14× Lite usage”.
- Alibaba Token Plan Lite: 1x (the baseline). The 1x baseline. The lowest-priced Alibaba Token Plan; the higher tiers are compared by their monthly Credits. From “20,000 Credits a month”.
- Alibaba Token Plan Essential: 2.2x Alibaba Token Plan Lite. PlanYield arithmetic on published figures. PlanYield arithmetic: 45,000 ÷ 20,000 Credits a month. From “45,000 Credits a month”; “20,000 Credits a month” (Alibaba Token Plan Lite row).
- Alibaba Token Plan Standard: 4x Alibaba Token Plan Lite. PlanYield arithmetic on published figures. PlanYield arithmetic: 80,000 ÷ 20,000 Credits a month. From “80,000 Credits a month”; “20,000 Credits a month” (Alibaba Token Plan Lite row).
- Alibaba Token Plan Pro: 10x Alibaba Token Plan Lite. PlanYield arithmetic on published figures. PlanYield arithmetic: 200,000 ÷ 20,000 Credits a month. From “200,000 Credits a month”; “20,000 Credits a month” (Alibaba Token Plan Lite row).

Computed multiples are PlanYield arithmetic on published figures, not measured usage. “Reported” multiples come from coverage or a vendor staff post, not the vendor’s page.

No published multiple: ChatGPT Free, Cursor Free, Devin Free, GitHub Copilot Free, Google Antigravity Free, Kiro Free, ChatGPT Go, Kimi Code Plus, Kimi Code Pro, Alibaba Coding Plan Pro, Kimi Code Max, Kimi Code Ultra, and Devin Max.

Alibaba’s plans stop at their caps: Coding Plan Pro has hard request caps with no overage, and Token Plan pauses once its Credits and any Extra Bundles are used ([Alibaba limits](https://planyield.dev/limits/alibaba-model-studio/#hard-caps)).

## What is not claimed

The calculator does not estimate hours, messages or tokens, and it does not say which plan lasts longer in a real week. For that, read or add community usage reports on the [Report usage](https://planyield.dev/report/) page.

More: [Scoreboard](https://planyield.dev/) · [Compare two plans with a verdict](https://planyield.dev/compare/plans/) · [All comparisons](https://planyield.dev/compare/) · [All decision pages](https://planyield.dev/best/) · [Cheapest paid plans](https://planyield.dev/best/cheapest-ai-coding-plan/) · [How to check your usage](https://planyield.dev/guides/check-usage/) · [Data (CSV/JSON)](https://planyield.dev/data/)

Dataset licence: [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/).

## FAQ

### Why does the AI coding plan calculator not use hours?

Vendors do not publish their plan limits in hours. Their pages give prices, usage multiples, credit counts, message estimates and reset windows, so the calculator filters on those and never turns hours of use into a plan. For what plans deliver in real use, see PlanYield’s community usage reports on the Report usage page (/report/); they are published once a plan has 5 moderated reports.

### How does the AI coding plan calculator match a usage level?

It uses only usage multiples that are in the plan data, each against that vendor’s own entry paid plan, which counts as 1x: Cursor Premium is 4x Cursor Plus, GLM Coding Plan Pro is 6x GLM Coding Plan Lite, Google AI Ultra $100 (Antigravity) is 5x Google AI Pro (Antigravity), and Claude Max is 5x Claude Pro per 5-hour session and per week. A plan matches Heavy (~5× entry) when its multiple is at least 5x, and Very heavy (~20× entry or more) at 20x. A multiple from one vendor does not convert into another vendor’s.

### Which AI coding plans have no usage multiple in PlanYield’s data?

For one person: ChatGPT Free, Cursor Free, Devin Free, GitHub Copilot Free, Google Antigravity Free, Kiro Free, ChatGPT Go, Kimi Code Plus, Kimi Code Pro, Alibaba Coding Plan Pro, Kimi Code Max, Kimi Code Ultra, and Devin Max. The calculator still lists them, marked “multiple not published”, instead of leaving them out. Not every multiple it uses is published by the vendor: ChatGPT Pro 100, ChatGPT Pro 200, and ChatGPT Pro 500 are reported, from a vendor staff post or coverage, and Alibaba Token Plan Essential, Alibaba Token Plan Standard, GitHub Copilot Pro+, Kiro Pro+, Alibaba Token Plan Pro, GitHub Copilot Max, Kiro Pro Max, and Kiro Power are PlanYield arithmetic on published figures.

### What is the cheapest AI coding plan at each usage level?

Light (entry plan): Alibaba Token Plan Lite at $6 a month. At about 5x its own entry plan, by tool: Cursor: Cursor Super at $100 (15x Cursor Plus); Claude Code: Claude Max at $100 (5x Claude Pro per 5-hour session and per week); Codex: ChatGPT Pro 100 at $100 (10x ChatGPT Plus, reported); Copilot: GitHub Copilot Max at $100 (13.3x GitHub Copilot Pro, PlanYield arithmetic); Antigravity: Google AI Ultra $100 (Antigravity) at $99.99 (5x Google AI Pro (Antigravity)); Kiro: Kiro Pro Max at $100 (5x Kiro Pro, PlanYield arithmetic); GLM Coding Plan: GLM Coding Plan Pro at $80 (6x GLM Coding Plan Lite); Alibaba Model Studio: Alibaba Token Plan Pro at $43 (10x Alibaba Token Plan Lite, PlanYield arithmetic). No multiple in PlanYield’s data reaches it for Devin and Kimi Code. At about 20x or more, by tool: Cursor: Cursor Ultra at $200 (50x Cursor Plus); Codex: ChatGPT Pro 200 at $200 (20x ChatGPT Plus, reported); Antigravity: Google AI Ultra $200 (Antigravity) at $199.99 (20x Google AI Pro (Antigravity)). No multiple in PlanYield’s data reaches it for Claude Code, Copilot, Devin, Kiro, GLM Coding Plan, Kimi Code, and Alibaba Model Studio. Prices are monthly; matches use published, reported, or PlanYield arithmetic multiples, as each result says. Each multiple is against that product’s own entry plan, so the heavier levels have no single cheapest plan across vendors.

### Which AI coding plans avoid weekly caps?

Alibaba Token Plan Lite, Alibaba Token Plan Essential, Alibaba Token Plan Standard, Kimi Code Plus, Kimi Code Pro, Alibaba Token Plan Pro, Kimi Code Max, and Kimi Code Ultra say there is no weekly limit. For ChatGPT Free, GitHub Copilot Free, ChatGPT Go, GitHub Copilot Pro, Kiro Pro, GitHub Copilot Pro+, Kiro Pro+, GitHub Copilot Max, Kiro Pro Max, and Kiro Power, the terms PlanYield records do not mention one. ChatGPT Plus says weekly limits may apply. With “Avoid weekly caps” those stay in the results, flagged. The other 23 plans for one person record a weekly limit and are left out.
