PlanLimits
Every published plan for the tracked AI coding tools, with prices and limits taken from official vendor pages. Which plan fits you?
| Tool | Plan | Price/mo (USD) | Usage limits | Overflow / pay-as-you-go | Models | Last verified | Source |
|---|---|---|---|---|---|---|---|
| Claude Code | Pro | $20 | At least 5x more usage per 5-hour session than Free. Limits reset on a rolling five-hour window, and paid plans also have weekly limits. No fixed message count; usage depends on conversation length, model, and features. Claude and Claude Code share one pool. Anthropic may also apply weekly or monthly caps or model and feature limits. | Wait for reset, upgrade to Max 5x, enable usage credits after the five-hour or weekly limit, or switch to a Claude Console account and buy API credits billed at standard API rates. A small capped wrap-up allowance can finish an in-progress Claude Code response up to once per weekly usage period and counts toward the weekly limit. | Opus, Sonnet, and Haiku included. Fable is listed as usage credits, not included usage. | 2026-10-18 | source |
| Claude Code | Max 5x | $100 | 5x the Pro plan's per-session usage allowance. Session limit resets every five hours. Weekly usage limit applies across all models and resets at a fixed account time each week. Anthropic may also apply weekly or monthly caps or model and feature limits. No fixed message count is published. | Upgrade to Max 20x, enable usage credits after the five-hour limit, the general weekly limit, or the separate Fable weekly limit, switch to Console API credits at standard API rates, or wait for reset. Monthly API credits included with Max cannot be used in Claude Code. Wrap-up allowance can apply each time a five-hour session limit is hit, within the weekly limit. | Opus, Sonnet, and Haiku included. Fable is 50% of weekly limits. | 2026-10-18 | source |
| Claude Code | Max 20x | $200 | 20x the Pro plan's per-session usage allowance. Session limit resets every five hours. Weekly usage limit applies across all models and resets at a fixed account time each week. Anthropic may also apply weekly or monthly caps or model and feature limits. No fixed message count is published. | Enable usage credits after the five-hour limit, the general weekly limit, or the separate Fable weekly limit, switch to Console API credits at standard API rates, or wait for reset. Monthly API credits included with Max cannot be used in Claude Code. Wrap-up allowance can apply each time a five-hour session limit is hit, within the weekly limit. | Opus, Sonnet, and Haiku included. Fable is 50% of weekly limits. | 2026-10-18 | source |
| Claude Code | Team Standard | $25 | 1.25x the Pro plan's per-session usage allowance, plus a weekly usage limit across all models. Five-hour session limit and weekly limits reset on the account's schedule. Limits are per member and are not shared or transferable. | Owners can pre-purchase usage credits so members keep using Claude, Cowork, and Claude Code after the seat limit, billed at standard API rates. Organization-wide and per-user monthly spend limits and auto-reload are available. Hitting a user's spend limit pauses usage credits until the end of the month. | Access to all available models. | 2026-10-18 | source |
| Claude Code | Team Premium | $125 | 6.25x the Pro plan's per-session usage allowance, plus a weekly usage limit across all models. Five-hour session limit and weekly limits reset on the account's schedule. Limits are per member and are not shared or transferable. | Owners can pre-purchase usage credits so members keep using Claude, Cowork, and Claude Code after the seat limit, billed at standard API rates. Organization-wide and per-user monthly spend limits and auto-reload are available. Hitting a user's spend limit pauses usage credits until the end of the month. | Access to all available models. | 2026-10-18 | source |
| Claude Code | Team for scientists Standard | $0 | Not published | Not published | Not published | 2026-10-18 | source |
| Claude Code | Team for scientists Premium | $15 | Premium seats add five times more usage and higher limits for long-running analyses. | Not published | Not published | 2026-10-18 | source |
| Claude Code | Enterprise | $20 | No plan or seat-level usage limits and no included token allowance. Every token in Claude, Claude Code, and Cowork is billed at standard API rates. Admins can set organization and user spend limits. | There is no included usage cap. Self-serve usage is credits purchased upfront. Sales-assisted usage is billed monthly in arrears. Admins can set spend limits at the organization and user level. | Everything in the Team plan. The page does not name model versions. | 2026-10-18 | source |
| Claude Code | Enterprise Standard (seat-based legacy) | Not published | Per-seat usage limits apply. The page does not state the cap. | Usage credit options remain until the organization migrates to usage-based billing. | Not published | 2026-10-18 | source |
| Claude Code | Enterprise Premium (seat-based legacy) | Not published | Per-seat usage limits apply. The page does not state the cap. | Usage credit options remain until the organization migrates to usage-based billing. | Not published | 2026-10-18 | source |
| Claude Code | Enterprise Chat + Claude Code (legacy) | Not published | No per-seat usage limits. Usage is consumption billed at API rates. | No included allowance to exceed. Usage is billed at API rates from consumption. The page does not publish a spend-cap amount for this legacy seat. | Not published | 2026-10-18 | source |
| Claude Code | API | Not published | Monthly spend caps: Start $500, Build $1,000, Scale $200,000. Custom tier has no monthly spend cap. Evaluation tier starts below standard limits; amounts are not listed. Messages API limits are also RPM, ITPM, and OTPM by model class. For most models, cache reads do not count toward ITPM. | At the tier spend cap, API usage pauses until 00:00 UTC on the first day of the next month unless a higher limit is granted. Requests return HTTP 429 with no retry-after header. A self-set spend limit returns HTTP 400 until it is raised. RPM, ITPM, or OTPM limits return HTTP 429. Claude Code workspace limits are checked separately and can return 429 with retry-after. | Claude Fable 5.1: $10 input / $50 output per MTok, Claude Mythos 5.1 (limited availability): $10 input / $50 output per MTok, Claude Fable 5: $10 input / $50 output per MTok, Claude Mythos 5 (limited availability): $10 input / $50 output per MTok, Claude Opus 5.5: $4 input / $20 output per MTok, Claude Opus 5: $5 input / $25 output per MTok, Claude Opus 4.8: $5 input / $25 output per MTok, Claude Opus 4.7: $5 input / $25 output per MTok, Claude Opus 4.6: $5 input / $25 output per MTok, Claude Opus 4.5: $5 input / $25 output per MTok, Claude Opus 4.1 (retired, except Bedrock and Google Cloud): $15 input / $75 output per MTok, Claude Opus 4 (retired, except Google Cloud): $15 input / $75 output per MTok, Claude Sonnet 5.5: $2 input / $10 output per MTok, Claude Sonnet 5: $2 input / $10 output per MTok, Claude Sonnet 4.6: $3 input / $15 output per MTok, Claude Sonnet 4.5: $3 input / $15 output per MTok, Claude Sonnet 4 (retired, except Bedrock and Google Cloud): $3 input / $15 output per MTok, Claude Haiku 5.5 (prompts up to 100,000 tokens): $0.10 input / $0.50 output per MTok, Claude Haiku 5.5 (prompts over 100,000 tokens): $0.50 input / $2.50 output per MTok, Claude Haiku 4.5: $1 input / $5 output per MTok, Claude Haiku 3.5 (retired, except Google Cloud): $0.80 input / $4 output per MTok | 2026-10-18 | source |
| Cursor | Free | $0 | Limited. Every tier includes a pool of AI usage shared across Cursor and Grok (within Grok: Chat, Imagine, Voice, and Bot). The allowance refreshes weekly. | If you reach your included usage, you can keep going with top-offs, or wait for your allowance to refresh. | Grok 4.7, Grok 4.6, Grok 4.5, Composer 2.5 | 2026-10-18 | source |
| Cursor | Plus | $8 | Base. One weekly pool for Cursor models (Grok 4.7, Grok 4.6, Grok 4.5, and Composer 2.5), shared across Cursor and Grok. Plus runs those models at fixed settings: Fast off, medium effort, and Grok 4.7 uses the 256k context window. | Top-ups add prepaid credit for Cursor models beyond the weekly pool at the same per-model rates. Third-party models are Premium and above only. You are charged when you buy credit or when Auto Top-up runs, not in arrears. If Auto Top-up is off, additional usage stops once the weekly pool and any credit already bought are used up. A monthly additional usage limit pauses additional usage until next month or until you raise it, even if credit remains. | Grok 4.7, Grok 4.6, Grok 4.5, Composer 2.5 | 2026-10-18 | source |
| Cursor | Premium | $30 | 4x Plus. One weekly pool for Cursor models (Grok 4.7, Grok 4.6, Grok 4.5, and Composer 2.5), shared across Agent in the editor, CLI, Cloud Agents, Grok Bot, and the Grok apps. The pool refreshes weekly. | Top-ups add prepaid credit. Third-party models such as Claude or GPT are charged from the first request and are not part of the weekly pool. Cursor models beyond the weekly pool are charged at the same per-model rates. You are charged when you buy credit or when Auto Top-up runs, not in arrears. If Auto Top-up is off, additional usage stops once the weekly pool and purchased credit are used up. A monthly additional usage limit pauses additional usage until next month or until you raise it, even if credit remains. | Grok 4.7, Grok 4.6, Grok 4.5, Composer 2.5 | 2026-10-18 | source |
| Cursor | Super | $100 | 15x Plus. Same weekly pool for Cursor models (Grok 4.7, Grok 4.6, Grok 4.5, and Composer 2.5), shared across the subscription and refreshing weekly, plus priority model access at peak times. | Top-ups add prepaid credit. Third-party models such as Claude or GPT are charged from the first request and are not part of the weekly pool. Cursor models beyond the weekly pool are charged at the same per-model rates. You are charged when you buy credit or when Auto Top-up runs, not in arrears. If Auto Top-up is off, additional usage stops once the weekly pool and purchased credit are used up. A monthly additional usage limit pauses additional usage until next month or until you raise it, even if credit remains. | Grok 4.7, Grok 4.6, Grok 4.5, Composer 2.5 | 2026-10-18 | source |
| Cursor | Ultra | $200 | 50x Plus. Same weekly pool for Cursor models (Grok 4.7, Grok 4.6, Grok 4.5, and Composer 2.5), shared across the subscription and refreshing weekly, plus earliest access to new products and models. | Top-ups add prepaid credit. Third-party models such as Claude or GPT are charged from the first request and are not part of the weekly pool. Cursor models beyond the weekly pool are charged at the same per-model rates. You are charged when you buy credit or when Auto Top-up runs, not in arrears. If Auto Top-up is off, additional usage stops once the weekly pool and purchased credit are used up. A monthly additional usage limit pauses additional usage until next month or until you raise it, even if credit remains. | Grok 4.7, Grok 4.6, Grok 4.5, Composer 2.5 | 2026-10-18 | source |
| Cursor | Teams | $30 | Each paid seat has its own weekly pool for Cursor models (Grok 4.7, Grok 4.6, Grok 4.5, and Composer 2.5), shared across Cursor and Grok for that person. Seats do not share weekly pools. Premium uses the Premium weekly pool; Super is 3.75x Premium; Ultra is 12.5x Premium. The pool refreshes weekly. A Free seat has no team usage allowance and draws from the member's personal plan, or the small free allowance if they have none. | Third-party model usage, and Cursor model usage after a seat's weekly pool runs out, is paid from a prepaid balance the whole team shares. Admins buy one-time top-ups or turn on Auto Top-up. Third-party models are public list API prices plus a Cursor Token Rate of $0.25 per million tokens. Cursor models beyond the pool are billed at their token rates with no Cursor Token Rate. Additional usage stops when the prepaid balance is empty or the team's monthly additional usage limit is reached, whichever comes first. The page states no interruption in service or quality. | Grok 4.7, Grok 4.6, Grok 4.5, Composer 2.5 | 2026-10-18 | source |
| Cursor | Enterprise | Not published | Pooled. One shared committed amount for the contract period instead of fixed per-user allocations. All team members draw from that pool. Usage is tracked cumulatively across the contract term, not month by month. | On Enterprise pooled usage accounts, member spend limits apply to total usage, not just on-demand. Admins set team-level and member-level limits. When a spend limit is reached, additional usage stops until the limit is increased or the next period starts. | Grok 4.7, Grok 4.6, Grok 4.5, Composer 2.5 | 2026-10-18 | source |
| Gemini CLI | Gemini Code Assist Standard | $22.8 | 1500 maximum requests per user per day, combined with Gemini Code Assist agent mode. One prompt may count as multiple model requests. Daily limits are aggregated across model families (for example, Pro, Flash). Also limited per user per minute; the per-minute count is not published. Local codebase awareness quota: 1,000,000 token context window. | Once the daily maximum is reached, no further Gemini CLI or agent-mode requests can be made to any model until the quota resets. Most quotas can be adjusted by requesting a quota increase in the Google Cloud console. | Gemini model family as determined by Gemini CLI | 2026-10-18 | source |
| Gemini CLI | Gemini Code Assist Enterprise | $54 | 2000 maximum requests per user per day, combined with Gemini Code Assist agent mode. One prompt may count as multiple model requests. Daily limits are aggregated across model families (for example, Pro, Flash). Also limited per user per minute; the per-minute count is not published. Local codebase awareness quota: 1,000,000 token context window. | Once the daily maximum is reached, no further Gemini CLI or agent-mode requests can be made to any model until the quota resets. Most quotas can be adjusted by requesting a quota increase in the Google Cloud console. | Gemini model family as determined by Gemini CLI | 2026-10-18 | source |
| Gemini CLI | Workspace AI Ultra | Not published | 2000 maximum requests per user per day | Not published | Not published | 2026-10-18 | source |
| Gemini CLI | Google Developer Program | Not published | Higher usage limits for Gemini CLI. Also lists access to Gemini Code Assist, Cloud and GenAI monthly credits, and use with a corporate Workspace account. | Not published | Not published | 2026-10-18 | source |
| Gemini CLI | Gemini API key (unpaid) | $0 | 250 maximum model requests per user per day. Model requests to Flash model only. | If daily request limits are hit, switch to pay-as-you-go with a paid Gemini API key or Vertex AI. With an API key you pay per token or call. | Flash model only | 2026-10-18 | source |
| Gemini CLI | Gemini API key (pay-as-you-go) | Not published | Quota varies by pricing tier. | You pay per token or call. This is described as the way to keep a workflow from stopping when a quota limit is reached. | advanced paid only models | 2026-10-18 | source |
| Gemini CLI | Vertex AI (Express mode) | $0 | Free for up to 90 days. gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite, and gemini-2.5-flash-image: 10 requests per minute. gemini-3-pro-image, gemini-3.1-flash-image, gemini-3.1-pro-preview, and gemini-3-flash-preview: dynamic throughput. | Enabling billing removes the 90-day limit and increases quotas. After that you pay only for what you use. | gemini-3-pro-image, gemini-3.1-flash-image, gemini-3.1-pro-preview, gemini-3-flash-preview, gemini-2.5-pro, gemini-2.5-flash-image, gemini-2.5-flash, gemini-2.5-flash-lite | 2026-10-18 | source |
| Gemini CLI | Vertex AI (pay-as-you-go) | Not published | Quota is governed by a dynamic shared quota system or pre-purchased provisioned throughput. | You pay based on model and token usage. Described as the flexible option when daily request limits or Gemini Pro quota are exhausted. | Not published | 2026-10-18 | source |
| Gemini CLI | Free Discontinued | $0 | 1000 maximum model requests per user per day. Also limited per user per minute; the per-minute count is not published. | On June 18, 2026, Gemini CLI stopped serving requests for people using it free of charge with Gemini Code Assist for individuals. | Gemini model family as determined by Gemini CLI | 2026-10-18 | source |
| Gemini CLI | Google AI Pro Discontinued | $19.99 | 1500 maximum requests per user per day. Also limited per user per minute; the per-minute count is not published. | On June 18, 2026, Gemini CLI stopped serving requests for Google AI Pro. | Gemini 2.5 Pro, Gemini 2.5 Flash | 2026-10-18 | source |
| Gemini CLI | Google AI Ultra Discontinued | $99.99 | 2000 maximum requests per user per day. Also limited per user per minute; the per-minute count is not published. | On June 18, 2026, Gemini CLI stopped serving requests for Google AI Ultra. | Gemini 2.5 Pro, Gemini 2.5 Flash | 2026-10-18 | source |
| GitHub Copilot | Copilot Free | $0 | An allowance of GitHub AI Credits (numeric amount not published) and 2000 code completions per month. | When included AI credits are exhausted: upgrade and pay only the price difference to the next tier, or set a budget for additional usage at 1 AI credit = $0.01 USD. Additional usage may be capped; pay already-consumed additional usage to continue. Otherwise wait until the next monthly reset. | Auto model selection only | 2026-10-18 | source |
| GitHub Copilot | Copilot Student | $0 | An allowance of GitHub AI Credits (numeric amount not published) and unlimited code completions. | Not published | Auto model selection only | 2026-10-18 | source |
| GitHub Copilot | Copilot Pro | $10 | 1,000 base GitHub AI Credits plus a 500-credit flex allotment (1,500 total per month). Base credits are used first. Included credits do not carry over and reset at 00:00:00 UTC on the first day of each calendar month. Code completions and next edit suggestions are not billed in AI credits and remain unlimited on paid plans. | When included AI credits are exhausted: upgrade and pay only the price difference to the next tier, or set a budget for additional usage at 1 AI credit = $0.01 USD. Additional usage may be capped; pay already-consumed additional usage to continue. Otherwise wait until the next monthly reset. | Claude Haiku 4.5, Claude Haiku 5.5, Claude Sonnet 4.6, Claude Sonnet 5, Claude Sonnet 5.5, Gemini 3.7 Flash, Gemini 3.8 Flash, GPT-5 mini, GPT-5.3-Codex, GPT-5.4, GPT-5.4 mini, GPT-5.6 Luna, GPT-5.6 Terra, GPT-6 Luna, Grok 4.5, Grok 4.6, Grok 4.7, Kimi K3, MAI-Code-1.1-Flash | 2026-10-18 | source |
| GitHub Copilot | Copilot Pro+ | $39 | 3,900 base GitHub AI Credits plus a 3,100-credit flex allotment (7,000 total per month). Base credits are used first. Included credits do not carry over and reset at 00:00:00 UTC on the first day of each calendar month. Code completions and next edit suggestions are not billed in AI credits and remain unlimited on paid plans. | When included AI credits are exhausted: upgrade and pay only the price difference to the next tier, or set a budget for additional usage at 1 AI credit = $0.01 USD. Additional usage may be capped; pay already-consumed additional usage to continue. Otherwise wait until the next monthly reset. | Claude Haiku 4.5, Claude Haiku 5.5, Claude Opus 4.8, Claude Opus 4.8 (fast mode) (preview), Claude Opus 5, Claude Opus 5.5, Claude Fable 5, Claude Fable 5.1, Claude Sonnet 4.6, Claude Sonnet 5, Claude Sonnet 5.5, Gemini 3.7 Flash, Gemini 3.8 Flash, Gemini 4 Argon, GPT-5 mini, GPT-5.3-Codex, GPT-5.4, GPT-5.4 mini, GPT-5.4 nano, GPT-5.5, GPT-5.6 Luna, GPT-5.6 Sol, GPT-5.6 Terra, GPT-6 Astra, GPT-6 Luna, GPT-6 Sol, GPT-6.1 Sol, Grok 4.5, Grok 4.6, Grok 4.7, Kimi K3, MAI-Code-1.1-Flash | 2026-10-18 | source |
| GitHub Copilot | Copilot Max | $100 | 10,000 base GitHub AI Credits plus a 10,000-credit flex allotment (20,000 total per month). Base credits are used first. Included credits do not carry over and reset at 00:00:00 UTC on the first day of each calendar month. Code completions and next edit suggestions are not billed in AI credits and remain unlimited on paid plans. | When included AI credits are exhausted: upgrade and pay only the price difference to the next tier, or set a budget for additional usage at 1 AI credit = $0.01 USD. Additional usage may be capped; pay already-consumed additional usage to continue. Otherwise wait until the next monthly reset. | Claude Haiku 4.5, Claude Haiku 5.5, Claude Opus 4.8, Claude Opus 4.8 (fast mode) (preview), Claude Opus 5, Claude Opus 5.5, Claude Fable 5, Claude Fable 5.1, Claude Sonnet 5, Claude Sonnet 5.5, Gemini 3.7 Flash, Gemini 3.8 Flash, Gemini 4 Argon, GPT-5 mini, GPT-5.3-Codex, GPT-5.4, GPT-5.4 mini, GPT-5.4 nano, GPT-5.5, GPT-5.6 Luna, GPT-5.6 Sol, GPT-5.6 Terra, GPT-6 Astra, GPT-6 Luna, GPT-6 Sol, GPT-6.1 Sol, Grok 4.5, Grok 4.6, Grok 4.7, Kimi K3, MAI-Code-1.1-Flash | 2026-10-18 | source |
| GitHub Copilot | Copilot Business | $19 | 1,900 GitHub AI Credits per user per month, pooled at the billing entity level. Included credits do not carry over and the pool resets at 00:00:00 UTC on the first day of each calendar month. Code completions and next edit suggestions are not billed in AI credits and remain unlimited. | Additional usage is enabled by default and continues at 1 AI credit = $0.01 USD, charged to the organization or enterprise. Additional usage may be capped; pay already-consumed additional usage to continue. If an admin disables the AI credits paid usage policy, usage is blocked until the next cycle. A user-level budget can halt that user even if the shared pool still has credits. | Claude Haiku 4.5, Claude Haiku 5.5, Claude Opus 4.8, Claude Opus 4.8 (fast mode) (preview), Claude Opus 5, Claude Opus 5.5, Claude Fable 5, Claude Fable 5.1, Claude Sonnet 5, Claude Sonnet 5.5, Gemini 3.7 Flash, Gemini 3.8 Flash, Gemini 4 Argon, GPT-5 mini, GPT-5.3-Codex, GPT-5.4, GPT-5.4 mini, GPT-5.5, GPT-5.6 Luna, GPT-5.6 Sol, GPT-5.6 Terra, GPT-6 Astra, GPT-6 Luna, GPT-6 Sol, GPT-6.1 Sol, Grok 4.5, Grok 4.6, Grok 4.7, Kimi K3, MAI-Code-1.1-Flash | 2026-10-18 | source |
| GitHub Copilot | Copilot Enterprise | $39 | 3,900 GitHub AI Credits per user per month, pooled at the billing entity level. Included credits do not carry over and the pool resets at 00:00:00 UTC on the first day of each calendar month. Code completions and next edit suggestions are not billed in AI credits and remain unlimited. | Additional usage is enabled by default and continues at 1 AI credit = $0.01 USD, charged to the organization or enterprise. Additional usage may be capped; pay already-consumed additional usage to continue. If an admin disables the AI credits paid usage policy, usage is blocked until the next cycle. A user-level budget can halt that user even if the shared pool still has credits. | Claude Haiku 4.5, Claude Haiku 5.5, Claude Opus 4.8, Claude Opus 4.8 (fast mode) (preview), Claude Opus 5, Claude Opus 5.5, Claude Fable 5, Claude Fable 5.1, Claude Sonnet 5, Claude Sonnet 5.5, Gemini 3.7 Flash, Gemini 3.8 Flash, Gemini 4 Argon, GPT-5 mini, GPT-5.3-Codex, GPT-5.4, GPT-5.4 mini, GPT-5.5, GPT-5.6 Luna, GPT-5.6 Sol, GPT-5.6 Terra, GPT-6 Astra, GPT-6 Luna, GPT-6 Sol, GPT-6.1 Sol, Grok 4.5, Grok 4.6, Grok 4.7, Kimi K3, MAI-Code-1.1-Flash | 2026-10-18 | source |
| Grok Build | Free | $0 | Limited. Shared AI usage pool across Grok and Cursor; within Grok the page names Chat, Imagine, Voice, and Bot. Allowance refreshes weekly. Grok Build row: Limited. | If you reach your included usage, you can keep going with top-ups, or wait for your allowance to refresh. | Grok 4.7 | 2026-10-18 | source |
| Grok Build | Plus | $8 | Basic. Shared AI usage pool across Grok and Cursor; within Grok the page names Chat, Imagine, Voice, and Bot. Allowance refreshes weekly. | If you reach your included usage, you can keep going with top-ups, or wait for your allowance to refresh. Paid plans include Grok API access billed separately at standard model rates; API usage does not draw from the plan usage pool. | Grok 4.7 | 2026-10-18 | source |
| Grok Build | Premium | $30 | 4x more usage compared to Plus. Comparison table: 4x Plus. Shared AI usage pool; allowance refreshes weekly. | If you reach your included usage, you can keep going with top-ups, or wait for your allowance to refresh. Third-party models in Cursor are billed separately. Grok API usage is billed separately and does not draw from the plan pool. | Latest Grok models | 2026-10-18 | source |
| Grok Build | Super | $100 | 15x more usage compared to Plus. Comparison table: 15x Plus. Shared AI usage pool; allowance refreshes weekly. | If you reach your included usage, you can keep going with top-ups, or wait for your allowance to refresh. Third-party models in Cursor are billed separately. Grok API usage is billed separately and does not draw from the plan pool. | Latest Grok models | 2026-10-18 | source |
| Grok Build | Ultra | $200 | 50x more usage compared to Plus. Comparison table: 50x Plus. Shared AI usage pool; allowance refreshes weekly. | If you reach your included usage, you can keep going with top-ups, or wait for your allowance to refresh. Third-party models in Cursor are billed separately. Grok API usage is billed separately and does not draw from the plan pool. | Latest Grok models | 2026-10-18 | source |
| Grok Build | Teams Premium | $30 | Each seat includes the same usage as the matching individual tier. | After a seat uses up its included weekly usage, any further first-party usage is Additional Usage, paid from additional usage credits bought with top-ups. | Not published | 2026-10-18 | source |
| Grok Build | Teams Super | $100 | Each seat includes the same usage as the matching individual tier. | After a seat uses up its included weekly usage, any further first-party usage is Additional Usage, paid from additional usage credits bought with top-ups. | Not published | 2026-10-18 | source |
| Grok Build | Teams Ultra | $200 | Each seat includes the same usage as the matching individual tier. | After a seat uses up its included weekly usage, any further first-party usage is Additional Usage, paid from additional usage credits bought with top-ups. | Not published | 2026-10-18 | source |
| Grok Build | Teams Free | Not published | Not published | Not published | Not published | 2026-10-18 | source |
| Grok Build | Business | $30 | Increased rate limits | Not published | Imagine, Voice, Grok 4.7 | 2026-10-18 | source |
| Grok Build | Enterprise | Not published | Pooled | Not published | Imagine, Voice, Grok 4.7 | 2026-10-18 | source |
| Grok Build | API | Not published | Per-model requests per second and tokens per minute by spend tier since 2026-01-01. grok-4.7 tier 0-4 RPS 150, 172, 208, 312, 500 and TPM 50M, 53M, 60M, 74M, 100M. grok-build-0.1 tier 0-4 RPS 37, 50, 75, 125, 208 and TPM 10M, 15M, 25M, 45M, 85M. Tiers: $0, $50, $250, $1,000, $5,000. Enterprise rate limits available on request. Exceeding a limit returns 429. | Prepaid credits are deducted for API consumption. With the default $0 invoiced billing limit, API requests are rejected once prepaid credits are depleted. If monthly invoiced billing is enabled, prepaid credits are used first and the rest is charged to the payment method at month end, until the invoiced billing limit. | Catalog includes grok-4.7 and grok-build-0.1. Grok 4.7 Fast is not on the public API. | 2026-10-18 | source |
| OpenAI Codex | Free | $0 | Codex is included. Usage limits vary by plan. Codex Cloud is not included. | Credit purchases are available only to a limited group of Free users. If shown, credits apply after included usage and do not raise the included limit. Purchase amounts and prices vary by account, region, and plan. Credits last 12 months and are non-refundable except where required. | GPT-6 Luna | 2026-10-18 | source |
| OpenAI Codex | Go | $8 | Codex is included. Usage limits vary by plan. Codex Cloud is not included. | Credit purchases are available only to a limited group of Go users. If shown, credits apply after included usage and do not raise the included limit. Purchase amounts and prices vary by account, region, and plan. Credits last 12 months and are non-refundable except where required. | GPT-6 Luna | 2026-10-18 | source |
| OpenAI Codex | Plus | $20 | Local-message estimates per five-hour period, not fixed message limits: GPT-6 Astra 5-45; GPT-6.1 Sol 15-160; GPT-6 Sol 15-150; GPT-6 Luna 350-3,000. Local messages and cloud chats share the allowance. Weekly limits may also apply. Cloud tasks may use more than local messages. Fast mode uses included subscription usage at 2.5x the Standard rate. ChatGPT Work shares this usage. ChatGPT for Excel shares it on Plus once that pricing is effective. | If a limit is reached during an active turn, the agent can finish that turn subject to fair-use limits. Plus users can buy credits to continue without upgrading. Extra local chats can use an API key at standard API rates. On purchased credits, Fast mode is 2x the Standard rate. | GPT-6 Astra, GPT-6.1 Sol, GPT-6 Sol, GPT-6 Luna | 2026-10-18 | source |
| OpenAI Codex | Pro 100 | $100 | No five-hour usage limit in Work and Codex. An included usage allowance still applies. Work and Codex share that allowance. Remaining allowance and reset times are in Settings → Usage. Ultrafast is not included. | Included usage is used first. After plan limits, Codex usage draws from the credit balance. Credits also cover ChatGPT Work, Word, Excel, and PowerPoint when those features are on the plan. Buy credits in Settings → Usage or Codex Usage & Billing. Automatic reload is available where shown. Credits last 12 months and are non-refundable except where required. | GPT-6 Astra, GPT-6.1 Sol, GPT-6 Sol, GPT-6 Luna | 2026-10-18 | source |
| OpenAI Codex | Pro 200 | $200 | No five-hour usage limit in Work and Codex. An included usage allowance still applies. Work and Codex share that allowance. Remaining allowance and reset times are in Settings → Usage. Ultrafast is not included. | Included usage is used first, then the credit balance. Credits support Codex, ChatGPT Work, Word, Excel, and PowerPoint where available on the account. Buy them in Settings → Usage or Codex Usage & Billing. Buying credits on Pro 200 does not unlock Ultrafast. Support cannot reset Codex limits. | GPT-6 Astra, GPT-6.1 Sol, GPT-6 Sol, GPT-6 Luna | 2026-10-18 | source |
| OpenAI Codex | Pro 500 | $500 | No five-hour usage limit in Work and Codex. An included usage allowance still applies. Work and Codex share that allowance. Ultrafast for GPT-6 Astra and GPT-6.1 Sol uses included usage first at 8x the Standard rate for the selected model, then credits at 6x the Standard credit rate. | Included usage is used first, then the credit balance. Among Pro plans, Ultrafast is only on Pro 500: it uses included usage at 8x the Standard rate, then credits at 6x the Standard credit rate for GPT-6 Astra or GPT-6.1 Sol. Credits can be purchased for eligible Codex usage. Support cannot reset Codex limits. | GPT-6 Astra, GPT-6.1 Sol, GPT-6 Sol, GPT-6 Luna | 2026-10-18 | source |
| OpenAI Codex | Business Standard | $25 | Standard Business local-message estimates per five hours, not fixed limits: GPT-6 Astra 5-45; GPT-6.1 Sol 15-160; GPT-6 Sol 15-150; GPT-6 Luna 350-3,000. Same ranges as Plus on this page. Local messages and cloud chats share the allowance. Weekly limits may also apply. Cloud tasks may use more than local messages. Fast mode uses included subscription usage at 2.5x the Standard rate. | Per-seat limits apply first. If the user exceeds a limit and the workspace has purchased credits, usage continues from the shared pool. Credits are optional. If no credits are available, the feature is blocked until an owner adds more. Owners can use automatic recharge. Business credits are valid for 12 months after purchase. | GPT-6 Sol, GPT-6.1 Sol, GPT-6 Luna, GPT-6 Astra, GPT-5.6 Sol, GPT-5.6 Terra, GPT-5.6 Luna | 2026-10-18 | source |
| OpenAI Codex | Business Premium | $125 | Premium seats have 5x the included usage of Standard seats and no 5-hour limit. This page's Standard local-message estimates per 5-hour window are GPT-6 Astra 5-45; GPT-6.1 Sol and GPT-6 Sol 15-150; GPT-5.6 Sol 10-100; GPT-5.6 Terra 25-200; GPT-5.6 Luna 250-2,000. Cloud chats can use more than local messages. Premium seats can use their full existing allowance for Astra. Standard seats only include limited Astra usage. | After included usage is exhausted, eligible usage, including additional Astra usage, can continue on purchased workspace credits if credits are available and spending controls allow it. Otherwise wait for the allowance to reset. Workspace credit purchases are optional. | GPT-6 Sol, GPT-6.1 Sol, GPT-6 Luna, GPT-6 Astra, GPT-5.6 Sol, GPT-5.6 Terra, GPT-5.6 Luna | 2026-10-18 | source |
| OpenAI Codex | Business legacy Codex seat | $0 | No included Codex allowance. Every Codex action draws on the workspace credit balance. If no credits are available, Codex is blocked until an owner adds credits. Owners and admins can set a monthly spending limit for legacy Codex seats or for a member; hitting that limit pauses Codex until the limit changes or resets. | There is no included pool. Usage stops when credits run out or a spending limit is hit. It resumes when an owner adds credits, changes the limit, or the spending limit resets. Purchased credits last 12 months. Owners can turn on automatic reload. Used credits are not restored when a member is removed. | GPT-6.1 Sol, GPT-6 Astra, GPT-6 Sol, GPT-6 Luna, GPT-5.6 Sol, GPT-5.6 Terra, GPT-5.6 Luna, GPT-5.3-Codex, GPT-5.2, GPT-Image-2 | 2026-10-18 | source |
| OpenAI Codex | Enterprise | Not published | Enterprise/Edu users with flexible pricing have no fixed rate limits; usage scales with credits. Enterprise and Edu plans without flexible pricing have the same per-seat usage limits as Plus for most features. Plus local-message estimates per five hours on this page: GPT-6 Astra 5-45; GPT-6.1 Sol 15-160; GPT-6 Sol 15-150; GPT-6 Luna 350-3,000. Weekly limits may also apply. If the agreement specifies usage-based billing in USD, the Enterprise USD rate card and the agreement apply instead of credit rates. | Credit-based Enterprise workspaces buy a shared credit pool at the contract level. There are no per-seat usage caps by default; admins can set spend controls by group. When the pool is exhausted, advanced features pause unless owners enable overages or buy more credits. An overage limit of 0 blocks new credit-using requests. Token-based Enterprise agreements are charged in USD at contracted rates. Credit expiration and seat true-ups follow the order form. | GPT-6.1 Sol, GPT-6 Astra, GPT-6 Sol, GPT-6 Luna, GPT-5.6 Sol, GPT-5.6 Terra, GPT-5.6 Luna, GPT-5.3-Codex, GPT-5.2, GPT-Image-2, GPT-Rosalind-Research | 2026-10-18 | source |
| OpenAI Codex | Edu | Not published | Enterprise/Edu users with flexible pricing have no fixed rate limits; usage scales with credits. Enterprise and Edu plans without flexible pricing have the same per-seat usage limits as Plus for most features. Plus local-message estimates per five hours on this page: GPT-6 Astra 5-45; GPT-6.1 Sol 15-160; GPT-6 Sol 15-150; GPT-6 Luna 350-3,000. Weekly limits may also apply. | Edu workspaces on credit-based flexible pricing use a shared credit pool purchased at the contract level. There are no per-seat caps by default. When the pool is exhausted, advanced features pause unless owners enable overages or buy more credits through the account team. Credit expiration is defined in the order form. | GPT-6.1 Sol, GPT-6 Astra, GPT-6 Sol, GPT-6 Luna, GPT-5.6 Sol, GPT-5.6 Terra, GPT-5.6 Luna, GPT-5.3-Codex, GPT-5.2, GPT-Image-2 | 2026-10-18 | source |
| OpenAI Codex | API Key | Not published | No included subscription allowance is stated. Usage is billed at API pricing. Codex is available in the CLI, SDK, or IDE extension. Cloud features such as GitHub code review and Slack are not included. | All usage is pay-as-you-go at API rates. There is no included pool to exceed. ChatGPT-signed-in users can also run extra local chats with an API key after they hit plan limits, charged at standard API rates. | Model availability follows the API models available to your key. | 2026-10-18 | source |
| OpenCode | Free | $0 | Free models include Big Pickle plus promotional models available at the time, with a quota of 200 requests/day. | Not published | Big Pickle, GLM-5.3-Flash Free, DeepSeek V4.1 Flash Free, Space Bunny Free, Beam Free, LL Alpha Free, LongCat 2.5 Preview Free, Exo Free, MiMo-V2.6-Flash Free, MiMo-V2.5 Free, Nemotron 3 Ultra Free, Nemotron 3.5 Lightning Free, Nemotron 3.5 Super VL Free, Jev 1.13 Free | 2026-10-18 | source |
| OpenCode | Zen | Not published | No fixed request, token, 5-hour, or weekly allowance. Pay-as-you-go per 1M tokens. You can set a monthly spend limit for the workspace and each member; the docs example is a $20 monthly usage limit. | Paid requests fail with HTTP 402 once the balance reaches zero. The balance never goes negative. Auto-recharge can buy more credits when the balance runs low. Credit purchases add a processing fee of 4.6% + $0.31. | GPT 6 Astra, GPT 6 Sol, GPT 6.1 Sol, GPT 6 Luna, GPT 5.6 Sol, GPT 5.6 Terra, GPT 5.6 Luna, GPT 5.5, GPT 5.5 Pro, GPT 5.4, GPT 5.4 Pro, GPT 5.4 Mini, GPT 5.4 Nano, GPT 5.3 Codex, GPT 5.3 Codex Spark, GPT 5.2, GPT 5.2 Codex, GPT 5.1, GPT 5.1 Codex, GPT 5.1 Codex Max, GPT 5.1 Codex Mini, GPT 5, GPT 5 Codex, GPT 5 Nano, Claude Fable 5.1, Claude Fable 5, Claude Opus 5.5, Claude Opus 5, Claude Opus 4.8, Claude Opus 4.7, Claude Opus 4.6, Claude Opus 4.5, Claude Sonnet 5.5, Claude Sonnet 5, Claude Sonnet 4.6, Claude Sonnet 4.5, Claude Haiku 5.5, Claude Haiku 4.5, Gemini 4 Argon, Gemini 3.8 Flash, Gemini 3.7 Flash, Gemini 3.6 Flash, Gemini 3.5 Flash, Gemini 3.5 Flash Lite, Gemini 3.1 Pro, Gemini 3 Flash, Grok 4.7, Grok 4.6, Grok 4.5, Grok Build 0.1, Muse Spark 1.3, Muse Spark 1.2, Qwen3.8 Max, Qwen3.8 Flash, Qwen3.7 Max, Qwen3.7 Plus, Qwen3.6 Plus, Qwen3.5 Plus, DeepSeek V4.1 Flash, DeepSeek V4 Pro, MiniMax M3, MiniMax M2.7, MiniMax M2.5, GLM 5.3 Flash, GLM 5.3, GLM 5.2, GLM 5.1, GLM 5, Kimi K2.5, Kimi K2.6, Kimi K2.7 Code, Kimi K3, Mistral Large 4, Jev 1.13, Jev 1.13 Free, Big Pickle, GLM-5.3-Flash Free, DeepSeek V4.1 Flash Free, Space Bunny Free, Beam Free, LL Alpha Free, LongCat 2.5 Preview Free, Exo Free, MiMo-V2.6-Flash Free, MiMo-V2.5 Free, Nemotron 3 Ultra Free, Nemotron 3.5 Lightning Free, Nemotron 3.5 Super VL Free | 2026-10-18 | source |
| OpenCode | Go | $10 | Monthly dollar allowance per model. Each model: 5-hour = 20% of the monthly limit, weekly = 50%, monthly = 100%. Token rates per 1M tokens match Go Plus. Go monthly limit and estimated requests (5h / week / month): GLM-5.3-Flash $60, 6,320 / 15,790 / 31,580; GLM-5.3 $15, 220 / 540 / 1,080; GLM-5.2 $60, 880 / 2,150 / 4,300; Kimi K3 $15, 110 / 250 / 490; Kimi K2.7 Code $60, 1,350 / 3,380 / 6,750; LongCat-2.0 $60, 11,400 / 28,600 / 57,200; LongCat 2.5 Preview Free unlimited (limited time); Step 5 Preview $60 (struck $15, 4x limited time), 1,720 (struck 430) / 4,280 (struck 1,070) / 8,520 (struck 2,130); Beam Free unlimited (limited time); MiMo-V2.6-Flash $60, 30,100 / 75,200 / 150,400; MiMo-V2.6-Pro $15, 3,250 / 8,150 / 16,300; MiMo-V2.5 $60, 30,100 / 75,200 / 150,400; MiMo-V2.5-Pro $15, 3,250 / 8,150 / 16,300; MiniMax M3 $60, 3,200 / 8,000 / 16,000; Muse Spark 1.3 Contributor $60, 45,300 / 113,300 / 226,600; Muse Spark 1.2 Contributor $60, 45,300 / 113,300 / 226,600; Qwen3.8 Max $15, 160 / 400 / 810; Qwen3.8 Flash $30, 5,400 / 13,500 / 27,000; Qwen3.7 Plus $60 for both ≤256K and >256K, 4,300 / 10,800 / 21,600; DeepSeek V4.1 Flash $80 off-peak and peak, 34,670 / 86,670 / 173,330; DeepSeek V4 Pro $15 off-peak and peak, 1,050 / 2,600 / 5,200; Hy4 preview $30, 1,350 / 3,380 / 6,770; Hy3 $60, 4,300 / 10,750 / 21,500; Space Bunny $30, 3,130 / 7,810 / 15,630; Grok 4.7 $15 for both ≤200K and >200K, 169 / 423 / 845; GPT 6 Luna $15 for both ≤272K and >272K, 4,230 / 10,560 / 21,130; GPT 5.6 Luna $15 for both ≤272K and >272K, 2,050 / 5,100 / 10,250; Claude Haiku 5.5 $15 for both ≤100K and >100K, 3,850 / 9,620 / 19,230; Gemini 4 Argon Contributor $60 (struck $30, 2x limited time), 440 (struck 220) / 1,080 (struck 540) / 2,120 (struck 1,060). Request counts are estimates; actual usage varies. Limits may change. | Without extra credits, requests stop at the usage limits. If Console credits are available, enable Use balance and Go falls back to the pay-as-you-go balance instead of blocking. | Grok 4.7, GLM-5.3-Flash, GLM-5.3, GLM-5.2, GPT 6 Luna, GPT 5.6 Luna, Claude Haiku 5.5, Gemini 4 Argon Contributor, Kimi K3, Kimi K2.7 Code, LongCat-2.0, LongCat 2.5 Preview Free, Step 5 Preview, Beam Free, MiMo-V2.6-Flash, MiMo-V2.6-Pro, MiMo-V2.5, MiMo-V2.5-Pro, MiniMax M3, Muse Spark 1.3 Contributor, Muse Spark 1.2 Contributor, Qwen3.8 Max, Qwen3.8 Flash, Qwen3.7 Plus, DeepSeek V4.1 Flash, DeepSeek V4 Pro, Hy4 preview, Hy3, Space Bunny | 2026-10-18 | source |
| OpenCode | Go Plus | $40 | Same cap rule as Go: 5-hour = 20% of the monthly dollar limit, weekly = 50%, monthly = 100%. Token rates per 1M tokens match Go. Go Plus monthly limit and estimated requests (5h / week / month): GLM-5.3-Flash $180, 18,960 / 47,370 / 94,740; GLM-5.3 $120, 1,760 / 4,320 / 8,640; GLM-5.2 $180, 2,640 / 6,450 / 12,900; Kimi K3 $60, 440 / 1,000 / 1,960; Kimi K2.7 Code $180, 4,050 / 10,140 / 20,250; LongCat-2.0 $240, 45,600 / 114,400 / 228,800; LongCat 2.5 Preview Free unlimited (limited time); Step 5 Preview $240 (struck $60, 4x limited time), 6,880 (struck 1,720) / 17,120 (struck 4,280) / 34,080 (struck 8,520); Beam Free unlimited (limited time); MiMo-V2.6-Flash $120, 60,200 / 150,400 / 300,800; MiMo-V2.6-Pro $60, 13,000 / 32,600 / 65,200; MiMo-V2.5 $120, 60,200 / 150,400 / 300,800; MiMo-V2.5-Pro $60, 13,000 / 32,600 / 65,200; MiniMax M3 $180, 9,600 / 24,000 / 48,000; Muse Spark 1.3 Contributor $120, 90,600 / 226,600 / 453,200; Muse Spark 1.2 Contributor $120, 90,600 / 226,600 / 453,200; Qwen3.8 Max $60, 640 / 1,600 / 3,240; Qwen3.8 Flash $90, 16,200 / 40,500 / 81,000; Qwen3.7 Plus $180 for both context tiers, 12,900 / 32,400 / 64,800; DeepSeek V4.1 Flash $160 off-peak and peak, 69,330 / 173,330 / 346,670; DeepSeek V4 Pro $60 off-peak and peak, 4,200 / 10,400 / 20,800; Hy4 preview $120, 5,400 / 13,520 / 27,080; Hy3 $240, 17,200 / 43,000 / 86,000; Space Bunny $120, 12,500 / 31,250 / 62,500; Grok 4.7 $60 for both context tiers, 676 / 1,692 / 3,380; GPT 6 Luna $60 for both context tiers, 16,920 / 42,240 / 84,520; GPT 5.6 Luna $60 for both context tiers, 8,200 / 20,400 / 41,000; Claude Haiku 5.5 $60 for both context tiers, 15,380 / 38,460 / 76,920; Gemini 4 Argon Contributor $240 (struck $120, 2x limited time), 1,760 (struck 880) / 4,320 (struck 2,160) / 8,480 (struck 4,240). Request counts are estimates. Limits may change. | Same as Go: requests stop at the limits unless Use balance is enabled, in which case usage falls back to Console pay-as-you-go credits. | Grok 4.7, GLM-5.3-Flash, GLM-5.3, GLM-5.2, GPT 6 Luna, GPT 5.6 Luna, Claude Haiku 5.5, Gemini 4 Argon Contributor, Kimi K3, Kimi K2.7 Code, LongCat-2.0, LongCat 2.5 Preview Free, Step 5 Preview, Beam Free, MiMo-V2.6-Flash, MiMo-V2.6-Pro, MiMo-V2.5, MiMo-V2.5-Pro, MiniMax M3, Muse Spark 1.3 Contributor, Muse Spark 1.2 Contributor, Qwen3.8 Max, Qwen3.8 Flash, Qwen3.7 Plus, DeepSeek V4.1 Flash, DeepSeek V4 Pro, Hy4 preview, Hy3, Space Bunny | 2026-10-18 | source |
| OpenCode | Black | Not published | Not published | Not published | Including Claude, GPT, Gemini and more | 2026-10-18 | source |
| OpenCode | Console workspace | $0 | No included usage allowance and no seat-based cap. The workspace pays for usage from shared prepaid credits. Owners and admins can set a monthly usage limit for the workspace and for each member. | Paid requests fail with HTTP 402 when the balance reaches zero. A credit agreement can switch the workspace to monthly invoicing; requests then fail with HTTP 402 when the credit limit is reached or an invoice is overdue. Go extra usage can be paid from the same credits. | Not published | 2026-10-18 | source |
| OpenCode | Enterprise | Not published | If you have your own LLM gateway, OpenCode does not charge for tokens used. | Not published | Not published | 2026-10-18 | source |
| Windsurf | Free | $0 | Light quota to code with agents; unlimited inline edits; unlimited Tab completions. | Wait until the next daily or weekly reset. | SWE-1.6 | 2026-10-18 | source |
| Windsurf | Pro | $20 | Daily and weekly usage quota covering Devin sessions, Devin CLI, and Devin Desktop. The daily allowance is more than 1/7 of the weekly allowance. | Pay-as-you-go on-demand credits for usage past the quota. Credits are prepaid, roll over month to month, never expire, and can be topped up or auto-reloaded. | Full model availability, including frontier models from OpenAI, Claude, Gemini, and SpaceXAI, plus leading open source models; SWE-2 (free for self-serve plans through 2026-10-15; now 75% off list price through 2026-12-31). | 2026-10-18 | source |
| Windsurf | Max | $200 | Weekly usage quota only, with no daily cap; significantly larger than Pro. Shared by Devin sessions, Devin CLI, and Devin Desktop. Includes everything in Pro. | After the included allowance, on-demand credits keep you working. Credits are prepaid, roll over month to month, never expire, and can be topped up or auto-reloaded. | Everything in Pro: full model availability, including frontier models from OpenAI, Claude, Gemini, and SpaceXAI, plus leading open source models, and SWE-2 (free for self-serve plans through 2026-10-15; now 75% off list price through 2026-12-31). | 2026-10-18 | source |
| Windsurf | Teams | $40 | Each full seat includes a daily and weekly usage quota equivalent to Pro. The daily allowance is more than 1/7 of the weekly allowance. Flex seats have no included quota and draw from the shared on-demand credit pool. Up to 200 members. | Shared on-demand credits fund flex-seat usage and full-seat usage past the included quota. Credits are prepaid, roll over month to month, never expire, and can be topped up or auto-reloaded. The team has one shared balance. | Everything in Pro: full model availability, including frontier models from OpenAI, Claude, Gemini, and SpaceXAI, plus leading open source models, and SWE-2 (free for self-serve plans through 2026-10-15; now 75% off list price through 2026-12-31). | 2026-10-18 | source |
| Windsurf | Enterprise | Not published | Billed in Agent Compute Units (ACUs). The included ACU count depends on the contract. Consumption scales with inference and the model selected. Local-agent tokens convert to ACUs at the models-page rates. Legacy credit-based enterprise uses monthly prompt credits that do not roll over; the default premium-model message costs 1 credit. | No public pay-as-you-go rate. If an organization hits its ACU limit, new work billed to it is blocked until an admin raises or removes the limit or the next monthly billing window resets usage. | Not published | 2026-10-18 | source |