Contents 12 sections
What is Grok and who makes it?
Grok is an AI assistant from xAI for following public discussion and drafting content. Its connection to X makes breaking news and social trends a particular focus.1,2,3,4
Vendor and launch
Grok has a reported model timeline that includes Grok 4, launched on July 9, 2025, and Grok 4.5, publicly launched on July 8, 2026.
Those dates mark model releases, not the launch of the overall service.5,6
Who Grok is built for
Grok suits work that draws on live X information or turns current discussions into drafts. Its X focus is less useful when a task has little to do with social trends or public conversation.2,3,4
- Researchers can follow live X discussions with Grok. They still need to verify posts before using them to support a finding.2,3
- Marketers and content creators can draft content around current topics with Grok. That workflow makes the most sense when an audience or campaign follows public discussion.2,3,4
- Social media managers can track X trends and draft responses with Grok. Its X connection matters less if their work centers elsewhere.2,3,4
Grok key features and models
Grok handles conversational writing, current-information retrieval, multimodal input, and coding assistance. Its models target different work: Grok 3 for everyday tasks, Grok 4 for complex reasoning, and Grok 4.5 for coding and tool use. A review of Grok 3 flags complex reasoning, customization, document analysis, and data visualization as limitations.2,3,4,6
In xAI’s published evaluation, Grok 4 reached 91.7% accuracy on AIME 2025 without tools7.
Grok models and context windows
A context window measures how much text a model can consider in a request, in tokens. The table pairs documented window sizes with stated model roles; more input capacity does not prove better answers.4,5,6
The original Grok 3 API model had a documented 131,072-token context window8.
The table lists Grok models by context-window size and documented task focus, starting with the smallest window. The largest exceeds the smallest by 868,928 tokens8.
| Model | Context window | Documented task focus |
|---|---|---|
| Grok 3 | 131,072 tokens | Data extraction, coding, and text summarization |
| Grok 4 | 256,000 tokens | Reasoning and complex tasks |
| Grok 4.5 | 500,000 tokens | Coding, agentic tool use, and configurable reasoning |
Table data: 4,5,6,8,9,10
Grok 4.5 does not have the largest listed window; the earlier Grok 4.3 has more capacity. A large window helps when a request contains substantial source material, but the model’s supported tasks matter too.6
Grok 4.3 supports 500,000 more context tokens than Grok 4.511.
- Grok 3 supports live web retrieval, image recognition, and image generation.
- Grok 4 Heavy runs multiple reasoning agents in parallel to produce and compare answers to complex tasks. Grok 4 Code focuses on generating and debugging code. Both have a different emphasis from general-purpose Grok 4.4,5
- Grok 4.1 emphasizes natural conversation and a consistent personality. xAI says users preferred it to the previous production model in nearly 65% of blind pairwise evaluations during a 2-week silent production rollout. That finding concerns preference, not coding or research accuracy.4,12
- Grok 4.5 offers low, medium, and high reasoning effort through its API, with high as the default. Documented tools include server-side web search, X search, and code execution, reflecting its focus on tool-driven work.6
- Grok-2 and Grok-2 mini were earlier beta models for chat, coding, and reasoning. They show the model lineage rather than the current task distinctions above.13,14
Search, drafting, images, and automation
Grok can answer questions, draft content, summarize X discussions, identify themes in current conversations, and cite sources found through web search. Its humor and sarcasm may not suit work that calls for a consistently formal tone.1,2,4,5
- Current-information search draws on web and X material for questions about developing events. Users can inspect retrieved sources, but an X discussion summary cannot verify the claims beneath it.2,4,5
- Image generation and Imagine turn text prompts into images, and Imagine supports video generation. Grok can analyze uploaded pictures; visual input and generated output serve different purposes.4,5
- Voice and file input take Grok beyond text conversations. Its documented inputs include voice, images, and files, alongside live web data.4
- Scheduled Tasks run recurring prompts and send results by email or app notification. Users need not enter a fresh prompt for every update.4
- Code execution includes a built-in Python REPL for running and testing code. Documented API workflows support agentic tool calling, where the model invokes tools during a task.4,15
How much does Grok cost?
Grok offers limited free access. A pricing report dated November 2, 2026 names The X Subscription as the default purchase path for new customers. Paid tiers share a usage pool rather than promising a fixed message count, so price alone cannot tell users how many tasks they can complete.
The X Subscription became the default on October 12, 2026. Existing SuperGrok, Cursor, and X Premium subscribers can keep legacy plans unless they switch.16
Free plan and its limits
Users can try those workflows, but the free tier does not promise unlimited use.4,16
Paid plans compared
Grok’s paid subscriptions vary by price, relative usage, and Grok Bot eligibility. The reported US list prices exclude sales tax; monthly prices assume monthly billing, while annual prices require annual billing.16
The X Subscription’s entry-level Plus tier lists for $8/month with monthly billing, before sales tax17.
The table compares The X Subscription tiers by US list price, usage relative to Plus, and Grok Bot access. Ultra includes 50 times the published usage of Plus17.
| Tier | Monthly billing | Annual billing | Usage and Grok Bot |
|---|---|---|---|
| Plus | $8/month | $80/year | Base paid usage tier; Grok Bot not included |
| Premium | $30/month | $300/year | 4x Plus usage; Grok Bot included |
| Super | $100/month | $1,000/year | 15x Plus usage; Grok Bot included |
| Ultra | $200/month | $2,000/year | 50x Plus usage; Grok Bot included |
Table data: 16,17
New team customers can mix Premium, Super, and Ultra seats at the corresponding per-seat prices shown above. Plus is unavailable for teams, and Enterprise uses custom pricing. Usage multiples compare tiers; they do not guarantee a prompt or output count.16
A Super seat on a new team subscription lists for $100 per seat per month17.
Prices have also been reported for Grok plans outside the new-customer lineup. The table keeps those plans separate from the current purchase path because they are not interchangeable tiers.4,5,16
The separately reported standalone SuperGrok price is $30/month with monthly billing18.
The table compares separately reported and legacy Grok-access plans, including their billing conditions. The legacy X Premium+ plan lists for $395/year with annual web billing in the United States19.
| Plan | Reported price | Condition or access |
|---|---|---|
| SuperGrok Lite | $10/month with monthly billing; $8.33/month billed annually | Separately reported Grok plan |
| SuperGrok | $30/month with monthly billing; $25/month billed annually | Earlier pricing report associates the monthly plan with Grok 4 access |
| Grok Business | $30/seat/month with monthly billing; $25/seat/month billed annually | Separately reported business plan |
| X Premium | $8/month with monthly billing | Higher Grok usage limits than the free tier |
| X Premium+ | $40/month with monthly billing | Includes SuperGrok access |
| SuperGrok Heavy | $300/month in an earlier pricing report | Legacy figure for exclusive Grok 4 Heavy access; not a new subscription tier |
Table data: 4,5,16,20
The X Subscription and separately reported plans differ in name and access terms, even when prices match. People keeping a legacy subscription should not assume the new-customer tiers describe their plan.4,16
For an existing US subscriber on monthly web billing, the listed legacy Premium+ renewal price is $40/month19.
Credits, usage limits and extra costs
Paid seats in The X Subscription share a weekly first-party usage pool across Grok apps, Grok on X, Cursor, and Grok Bot. Tasks draw different amounts from it; xAI does not publish a fixed message, token, or task allowance. Third-party Cursor models use prepaid credit from the first request. Eligible first-party use beyond the pool can draw from that credit too.16
Premium’s published weekly pool is four times Plus’s, rather than a fixed prompt allowance17.
The X Subscription offers extra usage credit starting at $5 on the web. Auto Top-up stays off unless enabled, and the additional-usage spending limit starts unset. Top-up credits generally expire one year after purchase, which teams should consider when budgeting beyond their included pool.16
Grok API use is billed separately from the subscription pool. The rates below are usage-based charges, not allowances included with a paid app seat.21
The original Grok 4 API input-token rate was $3 per 1 million tokens; that model was retired in May 20268.
The table compares documented API charges for tokens, tool calls, and storage. The Grok API charges $5 per 1,000 web search tool calls11.
| API item | Price | Billing condition |
|---|---|---|
| Grok 4 input tokens | $3.00 per 1 million input tokens | Standard rate for requests within a 128,000-token context window |
| Grok 4 output tokens | $15.00 per 1 million output tokens | Same standard-rate context condition |
| Grok 4.5 input tokens | $2.00 per 1 million input tokens | No batch discount at launch |
| Grok 4.5 cached input tokens | $0.50 per 1 million cached input tokens | Cached-input rate |
| Grok 4.5 output tokens | $6.00 per 1 million output tokens | Output-token rate |
| Grok 4.3 input tokens | $1.25 per 1 million input tokens | 20% batch discount reported for the model |
| Grok 4.3 cached input tokens | $0.20 per 1 million cached input tokens | Cached-input rate |
| Grok 4.3 output tokens | $2.50 per 1 million output tokens | Output-token rate |
| Grok 4.5 web search, X search, or code execution | $5 per 1,000 calls for each tool | Tool calls billed separately from tokens |
| Grok 4.5 file attachment search | $10 per 1,000 calls | Tool calls billed separately from tokens |
| Grok 4.5 collections search | $2.50 per 1,000 calls | Tool calls billed separately from tokens |
| File storage | $0.025/GiB/day | Storage charge |
| Collection storage | $0.10/GiB/day | Storage charge |
Table data: 5,6
A pricing review reports a higher Grok 4.5 API rate for requests exceeding 200,000 tokens of context, without quoting an amount. An earlier public API beta offered $25 in free credits per month through the end of 2024. That past offer is not a current subscription allowance.6,11,13
Does the Grok app cost money to use?
Limited Grok app use does not require a paid subscription. The Free tier provides daily access, while paid tiers offer more relative usage. Prices and access can vary by region and billing channel, so app-store and web purchases may carry different terms.2,16
A pricing guide describes a Grok Bot trial with usage credit for a seven-day window and no automatic switch to a paid plan. An App Store trial, by contrast, can renew as a paid subscription unless canceled. After a trial or promotion, costs depend on the plan, billing channel, prepaid usage, and separately billed API activity.16,21,22
Grok limits, privacy and commercial use
Grok offers memory controls, while model and image-tool access varies by plan, region, and policy. xAI’s Terms of Service and Privacy Policy govern its messaging service. Access to generated content should not be taken as blanket permission for commercial use.4,15,23
- Memory controls let Grok retain information across conversations by default, and users can turn memory off. Private mode keeps subsequent conversations out of memory. That matters for work that should not carry into later chats.4
- Prompt privacy calls for care with sensitive material: xAI’s privacy policy asks users not to put personal information in prompts. Memory settings control what carries between conversations; they do not override that guidance.4
- Image-tool access faces platform restrictions. Following investigations into harmful image generation, X limited image generation and editing to paying subscribers. It also added geoblocks for certain images of real people where that content is illegal. Subscription and regional eligibility can affect these workflows.4
- Grok Bot automation pauses for human intervention at CAPTCHAs, payments, or logins when it lacks credentials. Workflows involving those steps cannot run fully unattended.24
Grok integrations, apps and API
Grok is available through a standalone web interface, X, and dedicated iOS and Android apps. Coding and automation tools offer other ways in. An app chat, a browser-controlling agent, and an API request have different controls and usage allowances.4,5,15,21,24
- Grok’s web, X, and mobile apps give users direct access to the assistant. Voice chat works on web and mobile; mobile apps add camera input and Imagine. xAI launched the standalone iOS app to make Grok available beyond X.4,25
- Grok Bot runs separately named cloud agents that can converse, use tools, control a browser, and pass work between agents. Windows, macOS, iOS, and Android apps provide access. MCP-server connectors link external apps, while a virtual computer handles tasks without a connector. Users can require approval for selected connected-app actions, including sending email.24,26
- Grok’s Zapier connection provides actions across more than 10,000 apps. Cross-app workflows still depend on the actions each connected app makes available.24
- Grok’s coding integrations include Grok 4 Code in Cursor and Grok 4.5 as the reported default in the Grok Build coding-agent CLI. A review lists Grok 4.5 across Cursor plans and through OpenRouter, Vercel, Cloudflare, Snowflake, and Databricks Mosaic. Connecting a personal X account adds the applicable X subscription benefits.5,6,21
- Grok’s xAI API gives developers foundation-model access and supports OpenAI- and Anthropic-compatible SDKs. Beta documentation described function calling and custom system prompts. App features and API capabilities carry separate access terms and limits.13,15
- Grok’s Tesla integration appeared in a Grok 4 review as an initial US rollout through the 2025.26 software update for vehicles with AMD Ryzen processors. Those hardware and software requirements mean vehicle access is not a general Grok app feature.5
Grok pros and cons
Grok has documented strengths in finding and summarizing X content, along with reported limits in accuracy and video summarization. These findings come from a review of an earlier version, not a test of newer models.27
Pros
- X account and image search: An earlier review describes Grok finding X accounts and images. That makes it useful for research centered on material published on X.27
- Tweet and article summaries: The same review describes Grok summarizing tweets and articles. Readers can scan that material before checking the original source.27
Cons
- Math-answer consistency: The earlier review reports inconsistent answers to math questions. Anyone relying on a calculation should verify the result.27
- Direct video summarization: The reviewed version did not summarize videos directly. That finding does not establish a limit for every later Grok model.27
Grok for live X research and source verification
Grok can search live X posts, threads, and trends to give researchers current discussion as context. It helps identify what people are saying, not whether a circulating claim is true.4
A review reports Grok 3 analyzed 90 sources in 52 seconds
A published review says Grok 3 analyzed 90 sources in 52 seconds on an example task. That reported result guarantees neither the same speed elsewhere nor the reliability of those sources.3
How to verify claims drawn from X posts and web results
Grok can surface a claim from live X discussion. Verifying it means checking the material behind the summary.4
- Original post: The source post shows what someone said and whether Grok preserved the wording and context.
- Independent source: A primary document or independent report checks a claim beyond other X posts repeating it.
- Publication timing: Dates on the post and supporting material help show whether the claim still describes the current situation.
Grok can find discussions worth investigating. The underlying sources determine whether a finding is ready to cite.4
Grok 4.5 coding and agentic benchmarks
Grok 4.5 has reported results for agentic coding and tool use. Those benchmarks cover different tasks, so they do not add up to a single reliability score. A review reports an Artificial Analysis Intelligence Index score of 54 points, placing the model No. 4 of 168 models evaluated.6
A review reports 82% on Terminal-Bench v2.1
Grok 4.5 scored 82% on Terminal-Bench v2.1, according to a published review of agentic-coding results. The review also reports an output speed of 85.6 tokens/second. Neither result predicts performance on a particular team’s codebase.6
According to that review, Cursor CEO Michael Truell said many on his team had made the model a daily driver. Cursor co-launched it, so the comment describes reported team use rather than an independent benchmark.6
A review reports 33% on the τ³-Banking tool-use benchmark
Grok 4.5 scored 33% on the τ³-Banking agentic tool-use benchmark, ranking No. 1 of 28 models in the review’s results. Agentic tool calling lets a model select and use tools during a task. Its ranking compares evaluated models; its score does not indicate perfect benchmark performance.6
Grok 4.5 also received a reported 93% on GPQA Diamond, a scientific-reasoning benchmark. That measures something different from banking tool use. Teams considering automated workflows should weigh the tool-use result against their tasks, not substitute the scientific-reasoning score.6
Who should use Grok – and who should not?
Grok fits work centered on software development, complex reasoning, or automation across apps and websites. It is not a substitute for source-critical research, professional legal or medical advice, or a private enterprise knowledge system without further governance.5,15,24
Grok is a good fit if
- Software developers can consider Grok 4 Code, which targets software development. Teams should match subscription or API access to their coding workflow.5
- Researchers working on open-ended problems can consider Grok 4 Heavy for complex tasks. Its reasoning focus does not remove the need to verify conclusions that depend on sources.5,15
- Small teams automating administrative work can use Grok Bot for tasks in apps or websites, including those without a connector. They should plan for the human handoffs described above.24
- US federal agency buyers have a procurement path: a Grok 4 review says xAI models are available through the General Services Administration schedule. Agencies still need their own suitability and governance review.5,15
Grok is a good fit when its documented coding, reasoning, or app-automation workflows match a team’s needs. Access terms and source verification still matter.5,15,24
Look elsewhere if
- Source-critical researchers cannot treat Grok’s answer as a replacement for checking the underlying evidence. Its findings need verification.15
- Legal or medical professionals need qualified judgment, not an AI-generated answer in place of professional advice. Grok is not presented as a replacement in either field.15
- Organizations building a private enterprise knowledge system need governance beyond what Grok alone is presented as providing. They should define data and access requirements before using it for that purpose.15
Grok alternatives
A comparison source lists starting prices for five Grok alternatives: Claude, Gemini AI Pro, ChatGPT, Perplexity Pro, and Microsoft Copilot. This comparison covers price, not equivalent features or usage allowances.4
The table orders the five alternatives by listed starting monthly price and states each billing condition.
| Alternative | Listed starting price | Billing condition |
|---|---|---|
| Claude | $17/month | Billed annually |
| Gemini AI Pro | $19.99/month | Listed monthly starting price |
| ChatGPT | $20/month | Listed monthly starting price |
| Perplexity Pro | $20/month | Listed monthly price |
| Microsoft Copilot | $21/month | Listed monthly starting price |
Table data: 4
- Claude has the lowest listed monthly-equivalent starting price, but it requires annual billing. It is not the price-only choice for someone avoiding an annual commitment.4
- Gemini AI Pro has the next-lowest listed starting price. That figure alone cannot establish whether its features suit a particular Grok workflow.4
- ChatGPT shares its listed starting price with Perplexity Pro. Price alone does not separate them here.4
- Perplexity Pro has the same listed monthly price as ChatGPT. Choosing between them requires more than this price comparison.4
- Microsoft Copilot has the highest listed starting price in the group. It is not the lowest-cost option on the figures shown.4
Grok FAQ
Is Grok owned by Elon Musk?
xAI built Grok, and Elon Musk founded xAI in 2023. That identifies the company behind Grok; it does not establish Musk’s personal ownership of the product. Grok debuted in late 2023.4
What do people use Grok for?
People use Grok for general research, coding assistance, and creating or revising images for social, editorial, or concept work. Each calls for a different check: verify research claims against sources, and review generated code and images for the intended project.15
Sources
- ↑Grok AI Reviews: User Feedback (2026) https://google.com/goto?url=CAESeAHEIkXX2wy_jkXPSh54FK1DfeGUvBgTHd51YAybywZvVHFnk5qcw7Txj6kPtRHyYD2Z5amkvR9Ax9pj8dR6xOzdGWxbFvenWlUcPPwD9MJW3fI8h_9WtfxHx43-nWyhyoK2mx5mTLgT2WfP_mbd6cmQ55WzLOvffw&gsessionid=RpKwr_0_zVRYuCVU6Y6Yb5UNCFFL6_D76-oHL_ntiHcajAsB8-PeGQ
- ↑Grok Review Features Pricing and Best Uses https://google.com/goto?url=CAESnQEBxCJF13Ed3V8tWg2bsOQ3teLxXhl8ih1wCYSP1qIbejR0WL17VqEfhqDMsi8beaUpjMv9fUWkKGCeXq7un5_KDontD_SxabW6F0vWUan7tRGRKjd-MR6DaVxqn83JYj50_aPAxfeQ_QIdmb4q7pth_s0KUqkzVo8913qfKXDri75CnFnrUksS17qWGP0bMJnaHLNe7-A9dIZLOQF7&gsessionid=RpKwr_0_zVRYuCVU6Y6Yb5UNCFFL6_D76-oHL_ntiHcajAsB8-PeGQ
- ↑Grok 3 Review: AI Features, Strengths, and Limitations Explained – Geeky Gadgets https://google.com/goto?url=CAESbAHEIkXXCdhw2SMYQaVmyqsY8xj_YnTuTpX9FPpEgDmi5jQJeI2JCy2EbnbEYoXOjozXQ5GHiEz5CJaW5eDZJwqaZSWUwzKhfttMvg7hcZSyQJrEfJCXhXm163uFPWSIgu-jgZLrAk-9Yu2HfQ&gsessionid=RpKwr_0_zVRYuCVU6Y6Yb5UNCFFL6_D76-oHL_ntiHcajAsB8-PeGQ
- ↑Grok Cheat Sheet: A Complete Guide to Elon Musk’s Chatbot | eWeek https://google.com/goto?url=CAESdwHEIkXX6mn26FSvY_pvOkGslwIT2zl3-uf0Zc8HWoqWtyKWlw4mbwMXAiqzLv6ONRLR9ARIFqqXVhJVRMKV9N9uBJgvFEkYqphPRRAFoBwWmYAmb9Fa73jBAdJAwBroRGRiOQ-GGIkhJvcd_F7_z9cp037-bmA_&gsessionid=SCtszxD8FEnr4XeI0ty4kZDqmZZPzLHwmByq-d8boPURfC_T4AqJXg
- ↑Grok-4 Review: Pricing, Features, Tutorial & Competitors 2026 https://google.com/goto?url=CAESeAHEIkXX2mmSL4V7oHEdOwuTVF_igSr4OU2WZvgoMcAEA4m540AH6ZK3Wamum5ZZl7k0HAN-ENSyp-Hfzuh_7rE4DPXiBVt6iv3BZbcQuE3dicdQxxPUzZNKS-o_4GWU06XZabC2U0sbzOIJMmdBsN_bVUfDO5deFA&gsessionid=RpKwr_0_zVRYuCVU6Y6Yb5UNCFFL6_D76-oHL_ntiHcajAsB8-PeGQ
- ↑Grok 4.5 review: benchmarks, pricing, and the verdict | eesel AI https://google.com/goto?url=CAESZgHEIkXXlTBAT8yoqWdMy84ffLB_4WKzBFJGk5BmX7o0_cJZGAUng5fMz2hqqxICO_uE5xzNMietVjc4T1BTvxD0wTQ6m1kUuQMcr8T5X_EuhakQEumJOuwZINCHjgxrNVVtuFD3dA&gsessionid=RpKwr_0_zVRYuCVU6Y6Yb5UNCFFL6_D76-oHL_ntiHcajAsB8-PeGQ
- ↑xAI, Grok 4 Fast https://x.ai/news/grok-4-fast
- ↑xAI Management API documentation. ([docs.x.ai](docs.x.ai https://docs.x.ai/developers/rest-api-reference/management/auth
- ↑xAI Management API documentation and Oracle’s Grok 3 model documentation. ([docs.x.ai](docs.x.ai https://docs.oracle.com/en-us/iaas/Content/generative-ai/xai-grok-3.htm
- ↑xAI Grok 4.5 model documentation. ([docs.x.ai](docs.x.ai https://docs.x.ai/developers/models/grok-4.5
- ↑xAI API pricing and xAI’s November 4, 2024, API Public Beta announcement. ([docs.x.ai](docs.x.ai https://docs.x.ai/developers/pricing
- ↑xAI, Grok 4.1 announcement. ([x.ai](x.ai https://x.ai/news/grok-4-1
- ↑What Is Grok? Inside Elon Musk’s ‘Rebellious’ AI https://google.com/goto?url=CAESkQEBxCJF19UjK8dww2jSZq4vGW7fWbhsPQxISs5HEaSCYf5IhouYfRXqp1GAeL-w6xAy9MHvOyXFzQ1_VSX4459QSU7legDTgcZ6wL4GzRv7m-su-0qW1eCPI6j_xCN3L_p4T5d0kXZr0H2wWOFVQGnLIPiQjbtUUs2GfDy7ShfHhbMTdyeeHjRYo5PQvBp4HnnX&gsessionid=SCtszxD8FEnr4XeI0ty4kZDqmZZPzLHwmByq-d8boPURfC_T4AqJXg
- ↑xAI, Grok-2 Beta Release. ([x.ai](x.ai https://x.ai/news/grok-2
- ↑Grok: AI Chat Assistant | AI Dreamhub https://google.com/goto?url=CAESXQHEIkXXHuezTbukuTv-apR3_zbL_A5kkJ0hprBymKV0k6VDxDfbjAmsfjDiCfgsME1DBGHDVva7jJ4obTV9pikLMSQ4KK-cbJfzWpESoWi2GLa-NEKTwUxOBPvCRg&gsessionid=SCtszxD8FEnr4XeI0ty4kZDqmZZPzLHwmByq-d8boPURfC_T4AqJXg
- ↑Grok Pricing 2026: Free, $8 Plus, $30 Premium, $100 Super, $200 Ultra https://google.com/goto?url=CAESvgEBxCJF12O2o3ePRGh4sdAb888avXCTSgWwvNgF2QG0kupRuWhUP384usCdnKvevcYw-32lHmeUZ1tzAc-A3_fEQrsdERCykMbIY4mtprIj5_8m6XaRBn_xq-ytF-W2ht7NDQwGwYpOzrT-QB2kJxisyjyuz_CS_lyFy-OR5Wo08mTX04sQb12UqmDctM_gW0ls0ExbUyVZxoMmLR4pL1h_j5YgAQYcb_RqbihnX4r6WfBlPbcrS9C7n34ETmNa&gsessionid=L9USAXEeNHY664L34N0FlVi3tM6p9wF3NTstFFA87GHBYr4PkMijsw
- ↑xAI, The X Subscription pricing. ([docs.x.ai](docs.x.ai https://docs.x.ai/x-subscription/pricing
- ↑xAI Grok Bot pricing https://x.ai/bot?trk=public_profile__posts-text
- ↑X Help Center, About X Premium https://help.x.com/en/using-x/x-premium
- ↑Justin McKelvey, SuperGrok Pricing (2026). ([justinmckelvey.com](justinmckelvey.com https://justinmckelvey.com/blog/supergrok-pricing
- ↑Grok Plans 2026: New Subscriptions and Legacy Benefits https://google.com/goto?url=CAESfAHEIkXXT7ru8X0R6AdCmJxmqcs-spgo6VTxBGtX8DaGlwSCWOD-MVsSR8–P2psfk9BaToi7l03WtPKWeGKX1J_pgdSZ_9wsWXLUgF4RgYKfxp6SiN7nswgQ7oyDtHJizevhCreRepgUR-WAiBm53YAefXXZ85vacxV_EM&gsessionid=L9USAXEeNHY664L34N0FlVi3tM6p9wF3NTstFFA87GHBYr4PkMijsw
- ↑Grok Bot Pricing: $30, $100 or $200 a Month (Nov 2026) | SPIRITT https://google.com/goto?url=CAEScQHEIkXX2njF1hIoMViNq2Jv9OkLA-qQm4Od_iJrrfn8SwK_BqW2_0R2fX8QPYqlhqg7XcI1Xi791y1yA574xMIl9A-r4gQjdXgCn4EPFcoqZIVMACYrhacV0yxQ3hoOpies_holyKSoLcgSXO6ZEfAs&gsessionid=L9USAXEeNHY664L34N0FlVi3tM6p9wF3NTstFFA87GHBYr4PkMijsw
- ↑Grok https://grok.com/
- ↑What is Grok Bot? | Zapier https://google.com/goto?url=CAESXQHEIkXXxxE40gmAzVjiKB0guEpLzMcVdB-2xqtKtVgNBWNi8Jxib5469sJoWkpQkBSVGCtD9ePHWmKLWUXs-qMmTHN-C_wwAy5ZWl65GUsFaDWig5wzDSr2oa_gcA&gsessionid=SCtszxD8FEnr4XeI0ty4kZDqmZZPzLHwmByq-d8boPURfC_T4AqJXg
- ↑Grok introduces AI-powered features to wider audience | Digital Watch Observatory https://google.com/goto?url=CAESiwEBxCJF1zALTmcTeiJRIdKzRbu8ZLvlqfxBuDVriYW-P1aA4W-1sMFNW5pkEbSOazfzoK1de5cNJzSxpYx76vMXJx89YuL9U9r0–5-ysngihp1DA5Dr0shVxMO5NvLagw9UFO7jMmRAuv3ql6mqhpcs81cybmfeaaoS4UANqZRU6ghvseFsMhh2Gog&gsessionid=SCtszxD8FEnr4XeI0ty4kZDqmZZPzLHwmByq-d8boPURfC_T4AqJXg
- ↑Grokbot Price Drop: What the Cheaper Tier Opens Up for Agent Teams | MindStudio https://google.com/goto?url=CAESeQHEIkXXlF4MVsYnehU7Uzs6a-0vPKzO5nuNfks93KkuozLzhc6IXM1_UOOZyAKwajyrwTu-d5NgeuYy2sebfFZ1BJu6GH-LbNvb5xb-yKmzGme8fxyVcuJpGcddYo3ShG3CyIrmnNjuks0TYkhlDw2dFDpKLVsQjtw&gsessionid=L9USAXEeNHY664L34N0FlVi3tM6p9wF3NTstFFA87GHBYr4PkMijsw
- ↑Grok AI Unleashed: Uncensored Review and First Look! https://google.com/goto?url=CAESlgEBxCJF1w550vKja-kphWzbzgJTIX_-ktCqSnC5d2uX7RkpbX7XYWSZwANtgk1SbmOtn_E3QZdSslBHpjxL9R_nwBWfQnofkYuXUalbzkJVl-MG72O-McCuhb5cBRCKG-30AJyQ2k6AvdKlprPuxu1NUcO8KAL1Cqf20nLLtOTGkSvqdYHf6ECQ23A1zG1Xr2Hc_f0UV1o&gsessionid=RpKwr_0_zVRYuCVU6Y6Yb5UNCFFL6_D76-oHL_ntiHcajAsB8-PeGQ
Prices and terms checked at the source: November 2026.