AI Coding Plan Comparison 2026: Prices, Limits, and Features

An AI coding subscription can cost $20 a month and run out of included usage before the next billing date. Claude Code and Codex have account-based limits, GitHub Copilot meters agent work through monthly AI Credits, and Cursor shares a weekly AI allowance with Grok. Coding sessions, model choice, and extra usage charges all affect what the monthly fee actually buys.

The subscription also determines where you can code. GitHub Copilot works inside existing IDEs and through Copilot CLI. Cursor and Devin Desktop offer agent-based code editors, and Devin also provides cloud agents. Dedicated plans from GLM, QwenCloud, MiniMax, and other providers connect through compatible external coding agents.

In this post, we compared 16 AI coding subscription and token-plan families, including ChatGPT with Codex, Claude Code, GitHub Copilot, Cursor with xAI, Devin, and lower-priced multi-model options. The comparison shows which free tiers and paid plans include a coding agent, how each allowance resets, which external clients work with it, and when additional charges begin.

Last updated: October 25, 2026.

AI coding plans by budget and use case

AI coding plan prices and usage limits

Standard monthly prices in USD unless another currency is shown. Taxes, regional eligibility, and promotions can change the amount charged. Discounted annual equivalents are identified separately.

Free AI coding access before you pay

Chat subscriptions that include AI coding agents

These subscriptions bundle coding access with the provider’s consumer AI account. Chat and coding often draw from the account’s limits, credits, or billing rules.

ChatGPT with Codex

Official plans: Codex pricing and ChatGPT plan access

ChatGPT Free provides limited access to Codex, Go costs $8 per month, and Plus costs $20. Pro has three monthly tiers at $100, $200, and $500. Free and Go use GPT-6 Luna at Standard speed in the desktop app, subject to rollout. Plus includes GPT-6.1 Sol and GPT-6 Luna in Codex, and the $500 Pro tier includes Ultrafast mode.

Codex and ChatGPT Work share usage under the ChatGPT subscription. Plus includes the CLI, IDE extension, web access, and cloud code review. Usage varies with the model, project context, tool calls, and task size. Pro increases available capacity according to the selected tier.

Sign in with ChatGPT to use the subscription allowance. An OpenAI API key is billed separately by token and does not include subscription-based cloud integrations. Eligible plans can extend included usage with purchased ChatGPT credits.

Read More: OpenAI Codex Commands Cheat Sheet

Claude Pro and Max with Claude Code

Official plans: Claude Pro and Max pricing

Claude Pro costs $20 per month. Max 5x costs $100 and Max 20x costs $200. Claude Code and Claude chat share the account’s five-hour and weekly usage limits. The Max tiers provide larger allowances, with model-specific restrictions that can affect how much coding you can run.

Claude Code includes Opus and Sonnet models, with Opus 5.5 available for complex coding tasks. Fable models have different billing rules: Pro uses usage credits for Fable, and Max plans can use up to half their weekly limit for those models.

Max 5x also includes $100 per month in API credits, and Max 20x includes $200 per month. These API credits are distinct from the Claude Code subscription allowance. Additional usage credits can continue supported work after subscription limits are exhausted. Use your Claude account sign-in for included Claude Code usage.

Read More: The Ultimate Claude Code Resource List: Agents, Skills, Plugins & More

Google AI Pro with Antigravity

Official plans: Google AI plans

Google AI Pro costs $19.99 per month in the US. The $100 and $200 Google AI Ultra tiers provide approximately 5x and 20x the Pro-tier Antigravity usage allowances. Antigravity supports agent-based coding in its standalone desktop app, CLI, and browser, with tools for managing parallel coding agents.

Antigravity quotas depend on task complexity, model usage, and account tier. Google applies five-hour and weekly limits to eligible Gemini usage and separate limits for other available models. Pro and Ultra subscribers can purchase top-up AI credits after reaching an eligible Antigravity quota. Availability varies by account and country.

Read More: Antigravity CLI Cheatsheet: Commands, Setup, Shortcuts, and Workflows

xAI subscription with Grok Build and Cursor

Official plans: xAI Free, Plus, Premium, Super, and Ultra

xAI now uses one subscription for Grok, Grok Build, and Cursor. Free costs $0 with limited agent requests. Plus costs $8 per month, Premium $30, Super $100, and Ultra $200. Premium includes 4x the Plus usage pool, Super includes 15x, and Ultra includes 50x. These are relative allowances, not fixed counts of coding tasks.

Grok Build works in a terminal or headless CLI and supports project files, diffs, worktrees, subagents, skills, plugins, hooks, and MCP servers. Cursor includes its own editor and agent. Premium and higher tiers unlock cloud agents, automations, Origin code hosting, and Bugbot code review. Plus includes Cursor with Grok models. Models from other providers are available from Premium, with their usage billed separately at API list prices.

The included AI allowance refreshes weekly and is shared across Grok Chat, Build, Imagine, Voice, Bot, Grok API access, and Cursor. Extra usage can be purchased through top-ups. The shared allowance pays for eligible Grok activity in Cursor. Third-party model charges in Cursor do not come out of that included pool.

Read More: Grok Build: SpaceXAI’s Open-source Terminal AI Coding Agent

Mistral Pro with Vibe

Official plans: Mistral Free and Pro pricing

Mistral Free includes limited Vibe coding sessions. Pro costs $14.99 per month and increases access to Vibe in the CLI, IDE, and web under fair-use limits. The current Free plan does not list a fixed monthly API-credit benefit.

Pro includes $25.50 per month in API credits in Mistral’s current USD pricing. Paid usage beyond included Vibe limits can continue at API rates when pay-as-you-go credits are enabled. Vibe also supports remote coding tasks alongside local CLI and IDE sessions.

Read More: Mistral Vibe: Open-Source CLI Coding Agent and Claude Code Alternative

Kimi membership with Kimi Code

Official plans: Kimi membership pricing

Kimi offers five membership tiers in China: Go at ¥49 per month, Plus at ¥99, Pro at ¥199, Max at ¥699, and Ultra at ¥1,399. Kimi Code and Agent Swarm are included with Plus and higher tiers, but not Go. Agent Credits scale from 2x the Go baseline on Plus to 28x on Ultra; Max and Ultra also allow up to four parallel Agent tasks. Andante, Moderato, Allegretto, and Allegro are legacy memberships for existing subscribers.

Kimi Code shares monthly membership credits with Kimi Work, Kimi Claw, Deep Research, slides, and other member features. The new Plus, Pro, Max, and Ultra tiers have a rolling five-hour Kimi Code limit with no weekly cap. Legacy plans retain both five-hour and weekly Kimi Code limits. Extra Usage can continue eligible work through a purchased balance when enabled.

Read More: Kimi Code CLI: Open-Source AI Coding Agent with Skills & MCP Support

IDE and cloud-agent coding subscriptions

GitHub Copilot Free, Pro, Pro+, and Max

Official plans: GitHub Copilot plans and pricing

GitHub Copilot Free includes limited chat and agent usage, Copilot CLI access, and 2,000 code completions per month. Pro costs $10 per month, Pro+ costs $39, and Max costs $100. All three paid tiers include unlimited inline code completions and next-edit suggestions. Copilot works in supported editors and IDEs, GitHub, and the terminal. Paid plans also include access to Copilot cloud agents and code review.

GitHub AI Credits pay for chat, agent mode, cloud agents, Copilot CLI, and other metered features. Pro includes 1,500 Credits per month ($15 usage value), Pro+ includes 7,000 ($70), and Max includes 20,000 ($200). One AI Credit equals $0.01 in usage value. These totals include a base allocation and an additional flexible allowance that GitHub can change. Credits reset on the first day of each calendar month, regardless of the subscription renewal date. Model prices and token consumption determine how quickly Credits are spent. Paid plans can purchase more credits after the included balance runs out.

GitHub replaced premium-request billing with AI Credits for individual usage-based plans in June 2026. A limited group of existing annual subscribers can use the legacy premium-request rules until their legacy arrangement ends. GitHub’s plan documentation has the current usage details.

Read More: 10 Best CLI AI Coding Agents in 2026: Open-Source & Commercial

Devin Desktop and Devin Cloud (formerly Windsurf)

Official plans: Devin individual pricing

Devin Desktop is the new name for the Windsurf editor. Free includes a small agent allowance, limited model access, unlimited inline edits, and unlimited Tab completions. Pro costs $20 per month and adds access to frontier models and Devin Cloud agents. Max costs $200 per month and raises agent usage limits. Developers can use Devin Desktop and Devin CLI for local coding tasks and Devin Cloud for delegated cloud work on eligible plans.

Paid agent usage has daily and weekly refresh windows. The number of tasks depends on model choice, session length, and task complexity. Devin does not publish a fixed request count for every tier. Pro and Max subscribers can purchase additional usage at API prices after reaching the included limit. Existing Windsurf subscriptions retain their previous plan pricing when moving to Devin Desktop.

Devin Desktop details explain the Windsurf rename and how current installations update.

Dedicated AI coding plans for external coding agents

These plans typically use a dedicated API key or subscription endpoint in compatible coding agents such as Claude Code, Codex, OpenCode, Cline, Cursor, and OpenClaw.

GLM Coding Plan

Official plan: GLM Coding Plan pricing and access

GLM Coding Plan costs $18 per month for Lite, $80 for Pro, and $168 for Max. Annual billing reduces the effective monthly prices to $12.60, $56, and $117.60. Lite includes 2,000 Credits per rolling five hours and 10,000 per week. Pro includes 12,000 and 60,000 Credits, and Max includes 28,000 and 140,000. Both time windows apply. GLM-5.3 and GLM-5.3-Flash are available on every tier.

The GLM coding endpoint works with more than 20 agent clients, including Claude Code, Codex, OpenCode, OpenClaw, and Cline. Credits reflect input, cached input, output, and eligible MCP tool usage. Off-peak model calls consume half the standard Credits rate. Personal plans are intended for the subscriber’s approved interactive agent use, subject to the plan’s usage policy.

BytePlus ModelArk Coding Plan

Official plan: ModelArk Coding Plan

BytePlus ModelArk Coding Plan costs $10 per month for Lite and $50 for Pro. Lite has five-hour, weekly, and subscription-month quotas, with Pro providing five times the Lite allowance. Published Lite request estimates differ between official references: approximately 1,200–1,900 per five hours, 9,000–12,000 per week, and 18,000–24,000 per subscription month. These are illustrative request estimates, not guaranteed counts; actual usage varies by model and context length. The coding model catalog includes GLM-5.3-Flash, DeepSeek V4.1 Flash, and Dola-Seed-2.0 Pro, with Auto selection available.

The endpoint works with Claude Code, Cursor, Cline, Kilo Code, Roo Code, OpenCode, and other compatible agents. BytePlus stops plan usage when the quota is exhausted. Another package or account balance is not charged automatically.

QwenCloud Token Plan

Official plan: QwenCloud Token Plan Individual

QwenCloud Token Plan Individual now has four tiers. Promotional monthly prices are $6 for Lite, $10 for Essential, $18 for Standard, and $68 for Pro. Their monthly quotas are 11,500, 25,500, 45,000, and 180,000 Credits respectively. The old seven-day quota was replaced by a subscription-month allowance with no weekly quota limit.

The Credits pool pays for Qwen, GLM, DeepSeek, and other included models in compatible coding clients. Credits renew at the start of each 30-day subscription cycle. When the quota runs out, calls pause unless you upgrade or use a purchased Credit Pack. You can configure the personal key on multiple devices, but account sharing, production backends, and automated batch jobs are prohibited.

Tencent Cloud Token Plan

Official plan: Tencent Cloud Token Plan overview

Tencent Cloud uses a simpler monthly Credits structure. Lite costs $7 for 1,000 Credits, Standard $17 for 2,600, Pro $51 for 7,900, and Max $103 for 15,900. The model pool includes MiniMax, Kimi, GLM, and DeepSeek options.

The subscription works with Claude Code, OpenCode, Cline, Cursor, Kilo Code, Codex CLI, and more. Credits reset by subscription month. The personal plan is restricted to interactive use by the subscriber. Automation scripts, application backends, and shared accounts require another billing arrangement.

MiniMax M Plan

Official plan: MiniMax M Plan pricing and features

MiniMax M Plan costs $22 per month for Go, $55 for Explore, and $132 for Build at standard monthly rates. All three paid tiers include MiniMax Code for coding tasks, access to the M3.1 Flash text model, image and speech generation, web search, multimodal understanding, and a Subscription Key for compatible external coding agents. Explore includes three times the Go usage, and Build includes 7.5 times the Go usage. H3 video generation is included with Explore and Build, but not Go.

The M Plan Subscription Key works with external coding agents such as Claude Code, Codex, Cursor, OpenCode, and OpenClaw. Usage draws from the subscribed M Plan allowance. If that allowance runs out, an available purchased credit pack can fund eligible requests; the normal pay-as-you-go API account balance is not charged through a Subscription Key. The Free tier includes limited use of MiniMax Code and the M3.1 Flash model, but does not include an external-agent Subscription Key. The previous Plus, Max, and Ultra Token Plans are no longer sold to new subscribers; existing auto-renewing subscribers can retain their legacy terms or switch to M Plan.

Read More: Generate Text, Image, Video, Speech, and Music via CLI – MiniMax CLI

Xiaomi MiMo Token Plan

Official plan: Xiaomi MiMo Token Plan

Xiaomi MiMo Token Plan has four monthly tiers: Lite at $6, Standard at $16, Pro at $50, and Max at $100. Included monthly credits are 4.1B, 11B, 38B, and 82B respectively. Plans include MiMo V2.6 Pro and V2.6 Flash for compatible coding clients. MiMo V2.5 and V2.5 Pro reached their scheduled retirement date on October 21, 2026, at 10:00 a.m. Beijing time; their legacy model IDs have been retired.

MiMo Token Plan uses monthly credits rather than a separate five-hour or weekly subscription window. Requests from 00:00 to 08:00 Beijing time consume credits at 0.8x the regular rate. The Token Plan and Xiaomi MiMo Desktop Membership are different subscriptions with independent allowances.

Read More: MiMo Code: Free Open-Source AI Coding Agent with Persistent Memory

Multi-model AI coding subscriptions

Ollama includes one monthly cloud-usage credit balance that pays for its eligible models. OpenCode Go and Go Plus assign individual limits to each model across five-hour, weekly, and monthly windows. Actual request counts depend on each model’s token prices.

Ollama Pro and Max

Official plans: Ollama Free, Pro, and Max pricing

Ollama Pro costs $20 per month and includes $60 in monthly cloud-model usage credits. Max costs $100 and includes $300. Cloud usage follows the published per-token rates and no longer has the old session or weekly limits. Pro permits three concurrent requests, and Max permits ten.

After the included amount is exhausted, work can continue at per-token rates. Ollama also has a small Free allowance for starter cloud models. You can use it to test the hosted service and coding-agent connection before subscribing.

OpenCode Go and Go Plus

Official plan: OpenCode Go pricing and limits

OpenCode Go costs $10 per month. Go Plus costs $40 and has higher usage limits. Both work with OpenCode and other compatible coding agents. The model catalog includes selected GPT, Claude, GLM, Kimi, MiMo, MiniMax, Qwen, DeepSeek, and Grok models, with availability subject to change.

Each model has a separate usage budget. Five-hour and weekly allowances equal 20% and 50% of its monthly dollar-value limit. On Go, GLM-5.3-Flash has a $60 monthly cap, while Kimi K3.1 Pro has a $15 cap. The corresponding Go Plus limits are $180 and $60. Limited-time free models can be available separately, and purchased Zen credits can pay for eligible usage after a subscription limit.

AI coding plan usage limits and overage billing

Subscription login vs API key

A coding agent can charge an unexpected billing account when an old API key or environment variable overrides its subscription credentials. Codex uses ChatGPT sign-in for included subscription usage, and Claude Code uses a Claude account. xAI’s unified subscription shares a weekly AI usage pool across Grok and Cursor, including eligible Grok API use. GitHub Copilot and Devin meter eligible usage through their own product accounts. QwenCloud, MiniMax, and other dedicated plans require the correct subscription key and endpoint.

After changing credentials, restart the coding agent and send one small request. Check the subscription usage page and the API billing page before starting a long session. The subscription meter should record the request if the intended plan is active.

What happens when the quota runs out?

Quota behavior differs by provider. QwenCloud pauses at the monthly Credits limit unless you upgrade or have a Credit Pack. BytePlus stops at the plan limit. MiniMax M Plan uses the subscription allowance and can draw from purchased credit packs when that allowance is exhausted; its Subscription Key does not automatically charge the pay-as-you-go API balance. xAI refreshes its shared AI pool weekly and supports top-ups. Claude, Codex, GitHub Copilot, Devin, Ollama, Mistral, Google, Kimi, and OpenCode provide paid continuation options under the applicable account settings. In Cursor, third-party models are billed separately from xAI’s weekly Grok allowance.

Check the fallback setting before the first heavy coding session. A hard-stop plan keeps monthly spending predictable. A credit-based fallback keeps the agent running and can raise the total beyond the subscription price.

Can I share an AI coding plan?

Many individual coding plans are licensed for one subscriber and interactive use. QwenCloud, GLM, Tencent Token Plan, Kimi Code, and similar personal subscriptions restrict shared access, and some also restrict automated or backend workloads. Your team should use a plan that explicitly includes team seats, organization billing, and shared administration.

How to choose an AI coding plan

Start with a free option when it includes the editor or coding agent you plan to use. GitHub Copilot Free and Devin Free include editor-based coding, while xAI Free provides limited Cursor and Grok Build access.

For an IDE subscription, compare Copilot’s monthly Credits, Devin’s daily and weekly agent limits, and xAI’s weekly shared pool for Cursor and Grok. For a terminal agent or external client, check the compatible endpoints and the five-hour, weekly, or monthly quota that applies to the chosen plan.

Keep optional paid fallback disabled during your first billing cycle and use your own usage history to decide whether a larger tier is worth the extra cost.

FAQs

Does changing coding agents reset the quota?

No. The allowance belongs to the subscription account or plan key. Moving a GLM, QwenCloud, MiniMax, Kimi, MiMo, or OpenCode Go credential to another compatible agent continues to use the existing quota.

Can one coding agent use several subscriptions?

Yes, when the agent can configure multiple providers. Keep each subscription key paired with its correct base URL and provider profile. Every plan continues to use its own limits and billing account.

Does OpenAI-compatible mean every plan works in every agent?

No. OpenAI-compatible describes the API request format. The agent must also accept a custom provider, use a model ID the plan recognizes, and connect to the correct subscription endpoint.

Does DeepSeek have a monthly Coding Plan?

DeepSeek does not currently offer a fixed monthly Coding Plan. Its first-party API uses per-token billing. Several subscriptions in this comparison include DeepSeek models, including OpenCode Go, BytePlus ModelArk, QwenCloud Token Plan, and Tencent Cloud Token Plan.

Read More: How to Use DeepSeek in Codex and Claude Code

Related resources

• Claude Code Pricing and Usage Limits: Free, Pro, Max, Team, and API
• AI LLM API Pricing: GPT, Gemini, Claude & More
• 10 Best CLI AI Coding Agents: Open-Source & Commercial
• Gemini API Free Tier Limits: Requests Per Day & Setup
• OpenAI API Rate Limits: RPM, TPM, Tiers, and 429 Errors