Subscriptions, seats & agent plans — flat-rate AI pricing · ByteCosts
Subscriptions, seats & agent plans Flat-rate and per-seat plans (Claude, GitHub Copilot, Codex and more), plus self-hosted agent frameworks — deliberately kept separate from per-token API pricing . Filter by what actually includes coding or agent access.
Which plan should you pick? Individual coding Team coding Heavy / automation General chat & research
One developer using an AI coding assistant or agent day to day.
Best value
9/mo
Covers a typical individual coding workload at the lowest price.
Coding / agent access Renews per window · ~300/day Agentic coding, AWS console integration, transformation agent. pricing page /mo The biggest usage ceiling here — for individual coding at full tilt.
Coding / agent access Renews per window · ~5k/day Coding agent for all-day, high-volume use. pricing page Free pick
Free software
Free, open-source software with unlimited usage — you only pay the model provider.
Coding / agent access Unlimited usage (self-hosted) pricing page Picks are for a typical individual coding workload. “Premium interactions” count agent calls / messages / premium requests — providers meter differently and base autocomplete is often unmetered, so treat usage figures as directional. Copilot allowances are official; Claude figures are estimated from Anthropic's stated 5-hour windows and 5×/20× multiples. Browse every tier below.
All plans by product browse every tier in full Has free tier 44 products Codex agent available with Free limits (CLI, web, mobile).
GPT-5.3 with limited messages per window.
~10 messages / 5 hours
Includes Codex with Go-tier limits.
Higher limits than Free; global since Jan 2026.
Expanded message limits
Includes Codex coding agent (CLI, cloud, IDE).
Advanced models, Deep Research, weekly Codex usage.
Weekly Codex limits
Includes Codex for teams.
Shared team workspace with admin and SSO.
Per-seat limits · Admin console, SSO
5/seat billed annually.
Higher Codex usage: CLI, cloud, IDE.
5x Plus limits; 10x Codex promo through May 31, 2026.
5x Plus usage · 10x Codex (promo)
Intro tier; 10x Codex promo ends May 31, 2026.
Highest Codex usage: CLI, cloud tasks, IDE.
20x Plus limits for continuous, demanding workflows.
20x Plus usage · Highest Codex usage
Codex with enterprise controls.
Custom pricing; pooled usage, SSO, controls.
Negotiated usage
Custom contract.
Can run code; Claude Code is not included on Free.
Consumer chat and coding access with limited usage.
Usage limits apply · Web, mobile, and desktop
Includes Claude Code and Claude Cowork.
Everyday plan with more usage than Free.
Shared across Claude and Claude Code
00/year billed annually.
Team Standard Seat
5/mo
$20/mo annual
Includes Claude Code and Claude Cowork.
Team seat with more usage than Pro; SSO, central billing, admin.
Per-seat usage limits · Connectors and admin
0/seat billed annually.
Includes Claude Code through the Max family.
5x more usage per session than Pro.
Independent Sonnet and Opus buckets · Higher output limits + priority
Includes Claude Code and Claude Cowork.
Team seat with ~5x the usage of a standard seat.
Higher per-seat usage · SSO and admin
00/seat billed annually.
Includes Claude Code through the Max family.
20x more usage per session than Pro.
Independent Sonnet and Opus buckets · Highest limits + priority
AI code editor with limited agent.
Limited completions and agent requests.
Limited agent + completions
Full coding agent, Tab completion, background agents.
0 credit pool, frontier models, cloud agents, MCP.
6/month effective annually.
Team coding agent with central controls.
Per-user seats; SSO, privacy mode, analytics, admin.
Per-seat usage · SSO and admin
Coding agent for daily heavy use.
3x the usage credits of Pro.
~3x Pro usage credit
Coding agent for all-day, high-volume use.
20x usage and priority feature access.
~20x Pro usage credit · Priority features
Enterprise coding agent with pooled usage.
Pooled usage and dedicated support.
Negotiated pooled usage
Custom contract.
GitHub Copilot curated
6 plans
pricing Completions, chat, and coding agent (limited).
50 chat/agent + 2,000 completions / month.
50 premium requests / month · 2,000 completions / month
Coding agent, code review, chat, CLI, MCP.
Unlimited completions + 300 premium requests / month.
300 premium requests / month · Unlimited base completions
00/year billed annually.
Team coding agent with org management.
Org policy controls +
9 AI credits per seat.
9 monthly AI credits / seat · Policy, audit, IP indemnity
Coding agent, GitHub Spark, premium models.
Unlimited completions + 1,500 premium requests / month.
1,500 premium requests / month · All models + GitHub Spark
Enterprise coding agent + org knowledge bases.
Knowledge bases + $39 AI credits per seat.
$39 monthly AI credits / seat · Knowledge bases, enterprise features
Coding agent with the highest individual usage.
00 base credits +
00 flex allotment (
00 total).
00 base +
00 flex credits · Highest individual limits
Includes a flex allotment for overage.
AI IDE with Cascade coding agent (limited).
Windsurf Editor + Cascade agent with limited usage.
Limited agent usage
Full Cascade coding agent and model access.
Cascade agent + all models for daily use.
Monthly credit / usage allotment
Team Cascade coding agent with controls.
Per-user seats; admin, SSO/RBAC, central billing.
Per-seat usage · SSO/RBAC, admin
Cascade coding agent for all-day use.
High usage for heavy agent users.
High monthly usage allotment
Enterprise Cascade coding agent.
Custom; pooled usage and enterprise controls.
Negotiated usage
Custom contract.
StackBlitz-based app builder agent (limited).
Browser app builder with limited tokens.
Limited daily tokens
App builder agent with higher tokens.
Higher token allotment for app building.
Monthly token allotment
Team app builder agent.
Team collaboration and project management.
Per-seat tokens · Team collaboration
Enterprise app builder agent.
Custom token volume and controls.
Negotiated tokens
Custom contract.
Replit Agent app builder (limited).
Browser IDE + Agent with limited usage.
Limited monthly usage
Replit Agent build + deploy/hosting.
5 monthly credit, 5 collaborators, unlimited workspaces.
5 monthly credit · 5 collaborators
0/month effective annually.
Replit Agent with stronger models + private deployments.
00 credit, 15 collaborators, stronger models, private deploys.
00 monthly credit · 15 collaborators
$95/month effective annually.
Enterprise Replit Agent + deploy.
Custom; SSO, security, pooled usage.
Negotiated usage
Custom contract.
v0 app/UI generation agent (React/Next.js).
UI/app generation with $5 monthly credit.
$5 monthly credit
v0 app generation + Vercel deploy.
0 monthly credit, Vercel deploy, GitHub sync, visual edit.
0 monthly credit
Team v0 app generation + deploy.
$30 credit/user with shared team usage.
$30 credit / user
Business
Business v0 app generation + deploy.
Higher credit/user, business controls.
00 credit / user
AI code-review agent for pull requests.
AI PR review for open-source / individuals.
Open-source / individual use
AI code-review agent with team controls.
AI PR review with deeper analysis for teams.
Per-seat usage · 20% off annually
~
5/seat billed annually.
Enterprise AI code-review agent.
Self-host options, SSO, governance.
Self-host / SSO · Governance
Fully autonomous coding agent: plans, codes, tests, opens PRs.
Autonomous engineer; pay-as-you-go ACUs at ~
.25/ACU.
Pay per ACU (~
.25) ·
0 minimum
Usage-based ACU billing.
Autonomous coding agent for teams.
250 ACUs included at
.00/ACU; team controls.
250 ACUs included ·
.00/ACU thereafter
250 ACUs/month included.
Enterprise autonomous coding agent.
Custom ACU volume, SSO, controls.
Negotiated ACUs
Custom contract.
Gemini Code Assist curated
3 plans
pricing Gemini coding assistant/agent in IDE and CLI.
Individuals; very high daily completion and chat limits.
Generous daily code requests · IDE + CLI
Team Gemini coding agent across IDE, CLI, Cloud.
Per user; Google Cloud integration, higher limits.
Higher daily limits · Google Cloud integration
9/seat billed annually.
Enterprise Gemini coding agent with repo customization.
Per user; code customization, full GCP integration, governance.
Repo-aware customization · Enterprise governance
$45/seat billed annually.
Junie coding agent + AI Assistant (BYOK).
Junie coding agent in JetBrains IDEs with your own API key.
Pay your own model provider
AI Assistant and Junie coding agent.
AI Assistant + Junie with 10 AI credits / 30 days.
10 AI credits / 30 days
$8.33/month effective annually.
AI Assistant and Junie for regular coding-agent use.
AI Assistant + Junie with 35 AI credits / 30 days.
35 AI credits / 30 days
5/month effective annually.
Web app builder agent; custom domains, roles.
100 monthly credits + daily credits; custom domains.
100 monthly credits · Daily credits
Team web app builder agent.
SSO, team workspace, security center, internal publish.
Team workspace · SSO, security center
Enterprise web app builder agent.
Custom platform fee and controls.
Negotiated platform fee
Custom platform fee.
Coding agent for tests, review, and CLI.
Test generation, IDE review, CLI; limited usage.
Limited usage
Team coding agent: tests, review, CLI.
Test generation + PR review + CLI for teams.
Per-seat usage
Enterprise coding agent with governance.
Self-host/VPC, SSO, governance.
Self-host / VPC
Custom contract.
Amazon Q Developer curated
2 plans
pricing Agentic coding assistant with AWS integration.
AWS-focused IDE/CLI assistant with limited usage.
Limited monthly usage
Agentic coding, AWS console integration, transformation agent.
Higher limits; 4,000 LOC/user transformation pool.
4,000 LOC/user transformation · Overage $0.003/LOC
Transformation overage $0.003/LOC.
AI code completion and coding agent.
Code completion and chat; you-control deployment.
Per-seat usage
No free tier (retired).
Enterprise code assistant + agent with controls.
Privacy-first; self-host/VPC, policy, governance.
Self-host / VPC option · Policy and governance
AI chat/edit and inline assist in the Zed editor.
Native editor with limited AI credits.
Limited AI credits
AI chat/edit, inline assist, agentic editing.
Fast native editor with AI credits and model access.
Monthly AI credit pool
Codex software-engineering agent: edits, reviews, cloud tasks, worktrees.
Open-source Codex CLI; included in paid ChatGPT plans, or pay-as-you-go via the Codex rate card.
Pay per token (rate card) · No fixed seat fee
Usage-based; included with paid ChatGPT plans.
Assistant and Labs; not a dedicated coding agent.
Pro search with frontier models and higher daily limits.
300+ Pro searches / day · File and image analysis
00/year billed annually.
Perplexity Agent API curated
1 plan
pricing Programmatic agent/tool API (developer use).
Tool pricing for the Agent API, separate from model tokens.
Per-tool usage pricing
Usage-based tool pricing.
Developer
10x higher rate limits than free tier
Top open source model access with fast, high-context completions
24 million tokens/day
Top open source model for heavy coding workflows
120 million tokens/day
Highest rate limits for production workloads
Sim
Uso ilimitado. Proteção por hora: $0.15
Sim
Uso ilimitado. Proteção por hora: $0.60
Essential
Sim
Uso ilimitado. Proteção por hora:
.50
Sim
Uso ilimitado. Proteção por hora: $4.00
Automate coding, document analysis, and other tasks with open models
Light usage - Chatting with models, evaluating larger models, coding and AI assistants with smaller models
Access larger, more powerful cloud models; Run 3 cloud models at a time; 50x more cloud usage than Free
Day-to-day work - Larger models, coding automation, deep research; 50x more cloud usage than Free; 3 concurrent cloud models; 50x more cloud usage than Free
Max
Everything in Pro, plus 5x more usage than Pro
Heavy, sustained usage - Continuous agent tasks, multiple concurrent agents, large models over extended sessions; 5x more usage than Pro; 10 concurrent cloud models
Shared usage across your team; Centralized billing and administration; Single sign-on (SSO); Model access controls; MDM installer for Windows and macOS; Priority support and dedicated Slack channel
10 text prompts per day, 15 image prompts per day
Pro
Unlimited text prompts, 1,000 images per day, 100 credits/month for video, music, premium models, and API
Higher image generation limits, 7,500 credits/month for video, music, frontier image generation, LLMs, and API, 2-month credit banking
Highest image generation limits, 22,500 credits/month for video, music, frontier image generation, frontier LLMs, and API, 3-month credit banking
Unlimited autoscaling and priority compute access
Full control in your cloud and ours
Limited features including GitLab Duo Chat (agentic), Code Suggestions, Custom agents, External agents, Planner Agent, Data Analyst Agent, Developer Flow, Code Review Flow, Convert to GitLab CI/CD Flow, Fix CI/CD Pipeline Flow, Software Development Flow, MCP clients, SAST False Positive Detection Flow, SAST Vulnerability Resolution Flow, Security Analyst Agent
Limited to generally available features; beta and experimental features require credits in Monthly Commitment Pool
All features available on Free tier plus beta and experimental features
All generally available features including Custom flows, Initialize project context, Improve CI/CD setup, MCP server, CI Expert Agent, External MCP servers, Knowledge Graph, Resolve merge conflicts
All features available on Premium tier plus beta and experimental features
All features including all beta and experimental features
MiniMax Token Plan (minimax.io)
3 plans
pricing Yes, includes MiniMax-M3 for AI coding tools and agents.
MiniMax-M3: ~1.633B tokens/month; 5-hour rolling and weekly quota windows; supports roughly 3-4 agents; all API Platform models share quota.
Yes, includes MiniMax-M3 for AI coding tools and agents.
MiniMax-M3: ~5.053B tokens/month; 5-hour rolling and weekly quota windows; supports roughly 4-5 agents; all API Platform models share quota.
Yes, includes MiniMax-M3 for AI coding tools and agents.
MiniMax-M3: ~9.796B tokens/month; 5-hour rolling and weekly quota windows; supports roughly 6-7 agents; all API Platform models share quota.
MiniMax Token Plan (minimaxi.com)
3 plans
pricing Yes, includes MiniMax-M3 for AI coding tools and agents.
MiniMax-M3: ~600M+ tokens/month; 5-hour fixed and weekly quota windows; supports roughly 3-4 agents; all API Platform models share quota.
Yes, includes MiniMax-M3 for AI coding tools and agents.
MiniMax-M3: ~1.8B+ tokens/month; 5-hour fixed and weekly quota windows; supports roughly 4-5 agents; all API Platform models share quota.
Yes, includes MiniMax-M3 for AI coding tools and agents.
MiniMax-M3: ~5.5B+ tokens/month; 5-hour fixed and weekly quota windows; supports roughly 6-7 agents; all API Platform models share quota.
Explore
Tier 1 limits apply by default (Solar Pro 3: 30 req/sec, Information Extract: 10 req/sec)
Tier 2 limits apply (Solar Pro 3: 50 req/sec, Information Extract: 20 req/sec)
Tier 3 limits apply (Solar Pro 3: 80 req/sec, Information Extract: 40 req/sec)
Lite Plan
Vision Understanding, Web Search MCP, Web Reader MCP, Zread MCP
5-hour limit: approx. 80 prompts; Weekly limit: approx. 400 prompts; 100 web searches and web readers per month
Pro Plan
Vision Understanding, Web Search MCP, Web Reader MCP, Zread MCP
5-hour limit: approx. 400 prompts; Weekly limit: approx. 2,000 prompts; 1,000 web searches and web readers per month
Max Plan
Vision Understanding, Web Search MCP, Web Reader MCP, Zread MCP
5-hour limit: approx. 1,600 prompts; Weekly limit: approx. 8,000 prompts; 4,000 web searches and web readers per month
Coding Agent, Agentic extensions, image/video understanding, web search, web reading, open-source repository MCP
Approx. 80 prompts every 5 hours, approx. 400 prompts per week; 100 MCP calls per month (search, web reading, open-source repository)
Coding Agent, Agentic extensions, image/video understanding, web search, web reading, open-source repository MCP
Approx. 400 prompts every 5 hours, approx. 2000 prompts per week; 1000 MCP calls per month (search, web reading, open-source repository)
Coding Agent, Agentic extensions, image/video understanding, web search, web reading, open-source repository MCP
Approx. 1600 prompts every 5 hours, approx. 8000 prompts per week; 4000 MCP calls per month (search, web reading, open-source repository)
Cloudflare AI Gateway
2 plans
pricing 100,000 logs total across all gateways
10,000,000 logs per gateway, 10 million requests/month + $0.05/million
Cloudflare Workers AI
2 plans
pricing 10,000 Neurons per day free allocation
10,000 Neurons per day free allocation; $0.011 per 1,000 Neurons for usage above free allocation
Yes
10 requests per minute, 1,000 requests per day, 1,000,000 tokens per minute, 10,000,000 tokens per day
Yes
Custom - volume-based discounts, dedicated account manager, SLA guarantees, higher rate limits
Tencent Coding Plan (China)
2 plans
pricing AI coding scenario subscription
Every 5 hours: up to approximately 1,200 requests; Weekly: up to approximately 9,000 requests; Per subscription month: up to approximately 18,000 requests
AI coding scenario subscription
Every 5 hours: up to approximately 6,000 requests; Weekly: up to approximately 45,000 requests; Per subscription month: up to approximately 90,000 requests
AI coding tools including Claude Code, OpenClaw, OpenCode, Claude Code, Cline, Cursor, Qwen Code, Kilo Code, Kilo CLI
6,000 requests per 5 hours, 45,000 requests per week, 90,000 requests per month
Alibaba Coding Plan (China)
1 plan
pricing 支持所有编程工具（如 Claude Code、OpenClaw、Cursor、Hermes Agent、Qwen Code、Qoder、Lingma、Kilo CLI、Cherry Studio、Chatbox、Cline、Qoder Agent、OpenCode、Codex）
每5小时6,000次请求，每周45,000次请求，每月90,000次请求
Free / Pay‑as‑you‑go
Usage-based
None – pay per‑token/execution
Fire Pass (Early Access) $49/mo
Kimi K2.6 Turbo for personal agentic coding harnesses (Claude Code, OpenCode, Cline, Kilo Code, OpenClaw, LangChain Deep Agents)
Personal agentic coding use only; not for production workloads, team or shared usage
None — pay as you go
All available models
Pay-as-you-go access to Wafer-hosted models, billed per token
Pay-as-you-go access to Wafer-hosted models, billed per token
A_BETTER_YOU coding-agent runs
Daily metered usage. No monthly minimum, prepaid token bucket, or extra picker.
Agent frameworks self-hosted software — you pay the model provider separately O OpenClaw
free · open-source Self-hosted gateway for AI coding agents across chat apps and channel plugins.
Model billing: Model usage is paid to the selected provider, not to OpenClaw, unless a separate managed host is used.
docs H Hermes Agent
free · open-source MIT-licensed autonomous agent with native and OpenAI-compatible provider adapters.
Model billing: Supports Anthropic, Google, OpenRouter, Nous Portal, AWS Bedrock, DeepSeek, NVIDIA NIM, and other providers; provider billing applies separately.
docs Plans labelled curated are hand-verified from the provider's pricing page; the rest are extracted from official pages by the pipeline. Subscriptions bill a flat or per-seat rate and are independent of the per-token API costs in the provider pricing index . Agent frameworks are open-source/self-hosted — their model usage is billed by whichever provider you point them at.
Plan AI and cloud spend before it lands. Open the pricing index, then use the calculators to model your real workload.
For Engineering
Model costs by token, understand the economics of feature complexity.
For Finance
Budget forecasting and vendor negotiation with live pricing updates.
For Product
Compare models, simulate scenarios, monitor pricing changes in real time.
Browse tools Cost intelligence for AI, cloud, and SaaS. Public pricing, normalized into an index and calculators that engineering and finance can use in the same room.
Catalog: 137 providers · 4,993 models · updated Jun 1, 2026
Prices via models.dev and custom scrapers · model quality benchmarks via Artificial Analysis
Disclaimer: All information provided is for reference purposes only. Actual costs may vary based on usage patterns and provider terms. Always monitor your own token consumption and billing dashboard to track real expenses.
© 2026 ByteCosts. All rights reserved.
Built on public pricing data and browser-side calculators. Figures are directional.
