What to do when Grok Build hits its usage limit: 9 ways to keep coding in 2026

Disclosure: Enjoy is our product, and it runs Grok Build beside Claude Code and Codex. We checked every price and limit below against SpaceXSI’s, Cursor’s, Anthropic’s, OpenAI’s, OpenCode’s and Ollama’s own pages on November 10, 2026, and we say where another way is the better pick, including the free ones that don’t need Enjoy.

The ways to keep coding, compared

When a paid plan runs out, Grok Build stops and says “You hit your weekly limit.” Its source shows the three choices under that line: Upgrade tier, Buy more credits (“Purchase credits to keep using Grok Build”) and Try Again, which will “Resubmit the last prompt once you have usage again.” An account on the older on-demand billing sees “You’ve hit the credit limit for your plan.” or, with on-demand spending on, “You’ve hit your spending cap.” On the free plan the line is “You hit your free usage limit.”

The limit isn’t Grok Build’s own. Since June, Grok’s help says, paid plans have “one shared weekly usage pool that you can spend however you like across any Grok product,” in place of “separate daily limits for each product (like Chat, Imagine, Voice, or Build).” On The X Subscription, where new customers have started since October 12, the pool is wider: Cursor’s models page says it’s “shared across everything on your subscription: Agent in the editor, CLI, Cloud Agents, Grok Bot, and the Grok apps.” SpaceXSI doesn’t publish how big any pool is, only how the plans compare, and its usage limits page adds that “Usage rates vary by product, so different activities can draw different amounts from the pool.”

So start with two facts: what spent the pool, and when it resets. In Grok Build, /usage shows your weekly limit and the next reset. In Grok’s Settings → Usage, on the web and in the phone apps, you get “A percentage breakdown by product (API, Build, Chat, Imagine, Voice)” and “Your weekly reset date and time.”

Way	Cost, checked November 10, 2026	How soon you’re back	Best for
Wait for the weekly reset	Free	At the reset, up to a week away	A reset that’s a day or two off
Buy extra usage credits	From $5, spent at SpaceXSI’s “standard rates”	Right away	Finishing the task in front of you
Upgrade your plan	Premium $30 a month; Super $100; Ultra $200	Right away, on the web	Running out most weeks
An xAI API key	Per token: Grok 4.7 is $2 in and $6 out per million	Right away, once you sign out	Scripts, CI and a rare heavy day
Another model inside Grok Build	That provider’s API rates; nothing per token for a local model	Right away, in the same app	Staying in Grok Build on a model outside the pool
Claude Code	Claude Pro $20 a month, or $17 billed yearly; Max $100; Ultra $200	Right away, in a new conversation	A Claude plan you already pay for
Codex	Free with GPT-6 Luna in the desktop app, subject to rollout; ChatGPT Plus $20 a month; Pro from $100	Right away, in a new conversation	A ChatGPT plan you already pay for
OpenCode, with free or local models	Free	Right away, in a new conversation	No bill and no plan to run out
Enjoy	A monthly subscription, free for 7 days, with your phone and one computer (pricing)	Right away, in the same conversation when it began in Enjoy	Several labs’ agents, so the next limit doesn’t stop the work

Prices are each company’s US list prices, before tax. Waiting and OpenCode’s free models cost nothing extra; the other ways add to what you pay for Grok.

Wait for the weekly reset

Best for: a pool that resets in a day or two.

“The usage pool limit resets every week on a schedule shown in the Usage tab in Settings,” says Grok’s help, and until then “paid features will pause until your weekly limit resets.” Grok Build’s /usage shows the same reset, in your own time zone since a fix in its changelog. On The X Subscription, Usage at grok.com/dashboard “shows how much of your weekly pool you’ve used, the next reset date, and your prepaid credit,” per Cursor’s help. When the pool is back, Try Again resends the prompt that was stopped, and grok --resume reopens the session if you closed it.

Price: free.

Watch for: the reset can be most of a week away, and the free limits that remain are for other products: “You will still have access to Grok’s free tier limits on Chat and Voice.” On The X Subscription nothing carries over either: Cursor’s help says “Unused included usage does not roll over.” And the pool can empty while you aren’t coding. Grok Bot, which comes with Premium and above, “draws from that seat’s shared weekly usage pool, not a separate additive pool,” says SpaceXSI’s pricing page, and on November 1 many of its users reported their week’s usage jumping to 100%, as our memo explains.

One slower route to a refill: on Premium, Super and Ultra, a friend who becomes a paid subscriber through your Grok Bot referral link earns you “One free usage reset,” which “is added 7 days after your friend’s payment is received.” The referrer has to be “Based in the US,” and Cursor’s help says “The ability to use these resets is switching on gradually.” Neither page says in so many words what the reset refills, so don’t plan a deadline around it.

Buy extra usage credits

Best for: finishing the task in front of you today, without changing plans.

“Hitting the weekly usage limit is not a hard stop,” says Grok’s help: “You can purchase Extra Usage Credits to instantly continue usage.” Credits “are added to your account immediately,” they “are used after your plan’s included weekly usage is exhausted,” and “You can buy as little as $5 from the Usage tab on web.” On The X Subscription the same thing is called prepaid credit. Buy it under Additional usage credits at grok.com/dashboard/billing, where Auto Top-up “only runs after you turn it on and save it.” Grok Build’s Buy more credits choice opens Grok’s usage page on the web, and its changelog says “/usage now shows prepaid credits balance and auto top-up status.”

Price: from $5 a purchase. SpaceXSI doesn’t list a price per token for Grok Build’s credits. It says only that “Extra Usage Credits are priced at standard rates, which means the cost per action is higher than the effective rate you get from the included weekly usage in your plan.” For scale, its API lists Grok 4.7 at $2 per million input tokens and $6 per million output, and Cursor’s models page lists the same two figures for Grok 4.7 “when you buy additional usage beyond the weekly pool.”

Watch for: set a cap before a long run. The hard stop is a monthly additional usage limit, which SpaceXSI’s prepaid balance page calls a spending cap: “When you reach your spending cap, paid extra usage pauses until the next month or until you raise the cap, even if money remains in your prepaid credit.” Alerts aren’t one. Its spending caps page says “Alerts do not stop usage.” Credits “can currently only be purchased on the web,” and “Top up credits expire one year after purchase (unless otherwise stated).” On a team, “only admins can change automatic top-up settings.”

Upgrade your plan

Best for: running out most weeks, not once in a while.

SpaceXSI’s pricing sets Plus, at $8 a month, as the base. Premium, at $30, has “4x more usage compared to Plus,” Super, at $100, 15 times, and Ultra, at $200, 50 times. By those multiples, a Plus-sized week costs $8 on Plus, $7.50 on Premium, about $6.67 on Super and $4 on Ultra. Grok’s help makes the same case: “Upgrading to a higher plan provides significantly more weekly usage at a lower cost per action compared to repeatedly buying Extra Usage Credits.” On the web, “The higher tier starts immediately,” and Grok Build notices on its own: its changelog says “Subscription upgrades are now detected automatically without restarting the CLI.”

Price: Premium $30 a month, Super $100 and Ultra $200, or $300, $1,000 and $2,000 a year, per the plans’ price list.

Watch for: a bigger plan buys a bigger share of a pool SpaceXSI doesn’t size, and Premium and above add Grok Bot, which spends the same pool. The upgrade page doesn’t say whether the week you’ve already used is refilled. It says “The checkout screen shows how any remaining value is handled before you confirm,” so read that screen. Through Apple, “The higher tier starts after Apple confirms the change.” And if you’re on an older SuperGrok plan, moving to The X Subscription is a bigger step than an upgrade: “The switch is permanent. It can’t be undone, and Support can’t undo it for you.”

An xAI API key

Best for: scripts and CI, and a rare heavy day when paying per token beats a bigger plan.

Grok Build can run on an API key in place of your subscription. Its authentication guide says to “use an API key from console.x.ai” in XAI_API_KEY, and the API quickstart says to “Sign up for an account at console.x.ai, then load it with credits to start using the API.”

One step is easy to miss. “Grok uses the API key as a fallback when no session token is active. If you have already signed in interactively, the stored session token takes precedence. To fall back to the API key, run grok logout or delete ~/.grok/auth.json.” On a computer you keep, the enterprise page shows another way: “bind a key to a specific model in ~/.grok/config.toml.”

Price: per token, on SpaceXSI’s models page. Grok 4.7 is $2 per million input tokens, $0.50 for cached input and $6 per million output. grok-build-0.1 is $1 in and $2 out.

Watch for: it’s pay as you go from the credits you load, and “xAI API credits are non-refundable.” A long prompt costs double: “requests whose prompt reaches the listed token threshold are billed at the higher rate for all tokens in the request,” which for Grok 4.7 is $4 and $12 from 200,000 prompt tokens. Grok Build won’t show you the spend: its changelog says “/usage command and credit warnings are now hidden for API-key authentication,” so watch the Console. Grok 4.7 Fast “is not available on the public xAI API.” And check what your key bills to before a long run. Plus and above list “Grok API access,” and the Usage tab counts “API” among the products in the weekly pool, but SpaceXSI’s pages don’t say when a key draws on that pool and when on Console credits.

Another model inside Grok Build

Best for: staying in Grok Build, with your skills, hooks and sessions, on a model that isn’t in the pool.

“Grok supports any custom model,” say its docs. You add one to ~/.grok/config.toml with a base_url and the name of the environment variable that holds its key, then choose it with /model <name> or grok -m. The settings reference calls these “Custom / BYOK models (OpenAI-compatible or Anthropic Messages),” with api_backend set to chat_completions, responses or messages. That takes in Anthropic’s API, any OpenAI-compatible one, and a model on your own computer: Ollama, for one, “supports a subset of the OpenAI API” at http://localhost:11434/v1/.

Price: the provider’s API rates. Anthropic lists Claude Opus 5.5 at $4 per million input tokens and $20 per million output, and Sonnet 5.5 at $2 and $10. A local model costs nothing per token.

Watch for: the config takes an API key, so this is per-token billing with that provider. To use a Claude or ChatGPT plan instead, run Claude Code or Codex, below. Some of Grok Build’s tools belong to SpaceXSI’s servers: the setting supports_backend_search is for “Whether the endpoint supports Grok-hosted server-side search tools.” The reference also notes that “some BYOK endpoints need false” for stream_tool_calls. Update before you try it, because the changelog’s fixes from October include “Custom models on the Responses API with a custom base_url no longer fail every turn with an HTTP 400 schema error” and “Custom models on Chat Completions endpoints no longer fail turns with a tool schema error when MCP servers are connected.” What a local model can manage depends on your computer’s memory, which our local model guide covers.

Claude Code

Best for: a Claude plan you already pay for.

Claude Code is Anthropic’s coding agent, and it can pick up a project Grok Build has been working in, because the two read each other’s instructions. Grok Build’s project rules page says it reads “AGENTS.md, Agents.md, AGENT.md, CLAUDE.md, Claude.md, and CLAUDE.local.md,” and Claude Code “can also read a repository’s AGENTS.md files in place of CLAUDE.md.” To carry the conversation over, grok export <id> [file] will “Export a transcript as Markdown,” which you can hand to Claude as a file. The road back is shorter: grok import is there to “Import sessions from Claude Code.”

Price: Claude’s pricing marks Pro, Max and Ultra “Claude Code included.” Pro is $20 a month, or $17 a month billed yearly, Max is $100 and Ultra $200. The free plan leaves Claude Code out.

Watch for: Claude has limits of its own. Pro lists “More usage than Free, every 5 hours and every week,” though when a limit stops a task, “Claude Code waits in the open session and continues the task on its own after the limit resets.” Our Claude Code usage limit guide covers the rest. It starts a new conversation, so tell it where things stand, and it’s a second subscription beside Grok’s.

Codex

Best for: a ChatGPT plan you already pay for, or a free start.

Codex is OpenAI’s coding agent, in the ChatGPT desktop app, a CLI and an IDE extension. It starts from the same file Grok Build does: “Codex reads AGENTS.md files before doing any work.”

Price: Codex’s pricing gives Free and Go, which is $8 a month, “GPT-6 Luna at Standard speed in the desktop app, subject to rollout.” Plus, at $20, adds “Codex on the web, in the CLI, in the IDE extension, and on iOS” with GPT-6.1 Sol, and Pro is $100, $200 or $500.

Watch for: it can’t bring your Grok Build sessions with it. Its import guide says “The desktop app can import from Claude Code, Claude Cowork, or Cursor,” and the CLI from Claude Code or Cursor, so export the transcript as above and start a new conversation. On Plus, OpenAI estimates 15 to 160 local messages every five hours with GPT-6.1 Sol, but “These estimates are not fixed message limits,” and “Weekly limits may also apply.” “Pro plans currently have no five-hour limit.” Our Codex usage limit guide covers what to do when it runs out.

OpenCode, with free or local models

Best for: routine work with no bill and no plan to run out.

“OpenCode is an open source coding agent,” says its homepage, and “OpenCode comes with a set of free models. No account needed.” For a model on your own computer, its providers page says “Start Ollama and pull a model. OpenCode probes http://127.0.0.1:11434, discovers completion models, and adds them to /models without an account connection.” Our local model guide covers the setup and the memory it takes.

Price: free, with its free models or a model on your own computer. Its paid set, OpenCode Go, is “from $10/month.”

Watch for: the free models are temporary, and their terms differ. Each is listed as free or available “for a limited time.” Several say “The team is using this time to collect feedback and improve the model,” and one says “Its provider retains your data and may use it for model training.” It’s a different agent, so it starts a new conversation.

Enjoy

Best for: several labs’ agents in one place, so when Grok’s pool is empty, the same conversation carries on with another.

Enjoy is a desktop app for Mac, Windows and Linux that runs Grok Build, Claude Code, Codex, Antigravity, Muse Code, and local models through OpenCode, each on the plan you already have. When Grok Build runs out of usage in Enjoy:

• The conversation waits, then carries on. It shows Waiting for usage, and Enjoy “waits until the limit resets, then tries again, and the agent continues where it left off.”
• Or another agent picks it up. Select Stop waiting, choose Claude Code or Codex, and “The new agent gets the conversation so far, including replies from other agents, before it works on your message.” There’s no transcript to export.
• Usage in view. The agent settings show Grok Build’s usage as one bar, Weekly or Subscription, with when it resets.
• Work and personal accounts, side by side. Grok Build can keep several accounts signed in, and each project can use its own. By default, when one account runs out, Enjoy moves the work to another of yours that has usage left.

Price: Enjoy is one monthly subscription, and the first seven days are free, with a card to start. It includes your phone and any browser and one computer; each other computer is $9.99 a month, and each teammate you invite needs a subscription of their own. See pricing. Grok Build still runs on your Grok or X subscription, which SpaceXSI bills separately.

Watch for: Enjoy doesn’t refill Grok’s pool. Its help says “Enjoy doesn’t sell, buy or change plans,” so credits and upgrades still happen at grok.com. A second Grok account is a second subscription, and don’t count on one to get past a limit: SpaceXSI’s acceptable use policy bars “circumventing any rate limits or restrictions,” without saying whether a second paid account counts, and its terms say “You may not share your account credentials or make your account available to anyone else.” To keep Enjoy from moving work between accounts, set When usage runs out to Wait and resume. Grok Build “doesn’t tell Enjoy which account is signed in, so its accounts stay numbered,” and on Windows, saved accounts need Developer Mode outside WSL.

Waiting has conditions too: “Your computer must be on, with Enjoy open, when the usage resets,” and “Waiting doesn’t keep the computer awake.” Enjoy imports past Claude Code and Codex conversations, not Grok Build’s, so a session you began in Grok Build’s terminal starts over as a new conversation. Enjoy isn’t open source, it’s a monthly subscription on top of your AI plans once the free week ends, and scheduled workflows run only while your computer is awake with Enjoy open. If all you want is to wait, Grok Build’s own Try Again costs nothing.

When to stay on your plan and wait

If your week resets in a day or two, waiting is free. If you run out once in a while, a bigger plan is money spent for a rare afternoon, and Grok’s help agrees: “On lower tier plans, top ups are convenient for occasional extra usage.” Pay for that day with credits, and set a cap first. Upgrade when it happens most weeks.

Before you pay for more, find out what spent the last lot. Grok’s help says to “Use the product breakdown in the Usage tab to see your activity.” If Imagine, Voice or Grok Bot took the week, more Grok Build usage isn’t what you need. If Grok Build did, its own settings help the next week last:

• Skip the Fast model. Grok 4.7 Fast “costs 2x the standard token rates, or 1.5x for long-context requests,” and it’s “billed through your plan.”
• Cap long runs. /goal takes “--budget <tokens> to cap the tokens it may spend,” and --max-turns sets the “Maximum number of agent turns.”
• Think before fanning out. Subagents “are independent child sessions with their own context,” and --no-subagents turns them off for a session.
• See where a session went. The changelog says “/usage now shows token counts and cost for the current session.”

How we compared

We read SpaceXSI’s Grok help, X Subscription docs, pricing, terms and acceptable use policy, Grok Build’s docs, changelog and source on GitHub, and the xAI API’s pricing; Cursor’s models, usage and referral pages; Anthropic’s pricing and Claude Code docs; OpenAI’s Codex pricing, import and AGENTS.md docs and ChatGPT pricing; OpenCode’s homepage and docs; and Ollama’s docs, all on November 10, 2026, and every one is linked below. We quote each company’s own words for its limits and leave out what it hasn’t documented, such as the size of Grok’s weekly pool. We didn’t test these setups ourselves, and Grok Build’s limit message is quoted from its source code, not from a screenshot. Plans and limits change often, so we recheck this guide against its sources at least once a month, and the date at the foot of the page says when we last did.