Is Grok 2 Mini Worth the X Premium Subscription?

TL;DR

Grok 2 Mini performs well on logical reasoning, controlled-length summaries, marketing copy, and iterative coding, but its real-time answers are unreliable. Access requires an X Premium subscription priced at $8 per month, which also includes features such as verification, making the overall value reasonable for active X users despite inconsistent retrieval of current information.

Transcript

grock 2.0 was just released it's available right now inside of x.com this is the large language model from Elon Musk and X Ai and if you haven't used it before they had grock one then they had grock 1.1 this new version actually comes in two different sizes there is a version of it just called grock 2 and there's a version of it called grock 2 mini... Read More

Key Insights

• Grok 2 is xAI's strongest available model in the release, while Grok 2 Mini is the smaller option that was actually tested. Both models appear inside X, but the reviewer planned to evaluate the full model separately when access became available.
• Grok access requires an X Premium subscription priced at $8 per month. The subscription also includes other X benefits, such as a verified check mark, so the reviewer considered its overall value reasonable for people who already use the platform.
• Published chatbot rankings placed an early Grok version at number four, below ChatGPT 4o, Gemini 1.5 Pro, and another GPT 4o entry shown in the leaderboard. The benchmarks also presented Grok as competitive with Claude models, GPT-4 Turbo, and Llama 3.
• Grok 2 Mini handled logical reasoning correctly in the snail-well tests. It calculated 18 days for a 20-foot well and 28 days for a 30-foot well because the snail escapes before slipping backward on the final night.
• Grok 2 Mini followed a summary-length constraint accurately. When asked to condense a news article into 100–150 words, it returned 125 words and covered the Grok release, its reasoning and coding claims, leaderboard performance, and planned API availability.
• Fun mode changes Grok's conversational tone through a built-in toggle. In the fantasy role-playing game example, it delivered a playful response and recommended The Witcher 3, providing a stylistic option that would otherwise require an explicit tone prompt.
• Grok 2 Mini produced useful marketing copy with numbered points, bold formatting, and substantial product information. It also followed a second instruction to compress the copy into one sentence, showing that it could revise structure and length effectively through follow-up prompting.
• Real-time information was Grok 2 Mini's weakest tested capability. It sometimes surfaced recent X posts about Grok 2, but questions about the 2024 Olympics and Nvidia's current stock price produced incorrect, stale, or irrelevant answers, often requiring follow-up prompts.

Questions & Answers

Q: How well does Grok 2 Mini handle logical reasoning?

Grok 2 Mini correctly solved both logical reasoning tests presented in the review. For a snail climbing 3 feet each day and slipping 2 feet each night, it calculated 18 days to escape a 20-foot well. After the well was changed to 30 feet, it adjusted the answer to 28 days and correctly avoided subtracting the nightly slip after the final climb.

Q: Is Grok 2 Mini reliable for real-time information?

Grok 2 Mini was inconsistent when answering real-time questions. It initially claimed that the 2024 Summer Olympics had not occurred when asked which country won the most gold medals. It also returned an earlier Nvidia closing price and old X posts instead of the requested current stock price. A Grok 2 topic shortcut did retrieve posts from 11, 6, and 5 hours earlier.

Q: How much does Grok cost to use on X?

Grok required an X Premium subscription priced at $8 per month in the review. That payment did not cover only the chatbot, since the subscription also included other X features such as a verified check mark. The reviewer therefore considered the value fairly good for an active X user, especially because no usage limit was encountered during several hours of testing.

Q: What is the difference between Grok 2 and Grok 2 Mini?

Grok 2 is presented as xAI's best model in the release, while Grok 2 Mini is the smaller version. The reviewer only had access to Grok 2 Mini during the practical tests, even though both options appeared in the model selector. Because the stronger model was unavailable, the reviewer postponed a direct comparison with GPT 4o until Grok 2 could be tested.

Q: Can Grok 2 Mini summarize articles within a word limit?

Grok 2 Mini successfully followed the requested summary length in the demonstrated test. It was asked to summarize an entire news page in 100–150 words, and Microsoft Word counted the result at 125 words. The response covered the Grok 2 and Grok 2 Mini beta release, improved capabilities, chatbot leaderboard performance, and plans to make the models available through an API.

Q: Can Grok 2 Mini create a working checkers game?

Grok 2 Mini did not create a complete working checkers game on its first attempt. The initial version looked polished, but none of the gameplay worked. After being told that the game lacked logic, it generated a functional second version. That revision introduced a visual problem because all pieces appeared to share the same color, so another follow-up prompt was still needed.

Q: What does Fun mode do in Grok?

Fun mode gives Grok a more playful and expressive conversational style through a built-in toggle. When asked to recommend a fantasy role-playing game, it answered with theatrical language and suggested The Witcher 3. The reviewer liked having the tone available as a selectable mode because a comparable style would normally need to be requested directly in a chatbot prompt.

Q: Is Grok 2 Mini worth using instead of other chatbots?

Grok 2 Mini showed strong results in reasoning, summarization, marketing writing, formatting, and iterative coding, but it did not establish a clear overall advantage. Its expected strength, access to real-time X data, was unreliable in multiple tests. The $8 monthly subscription may still offer reasonable value to active X users because it includes Grok alongside verification and other platform features.

Summary & Key Takeaways

• Grok 2 and Grok 2 Mini were released through X by xAI, although only the Mini model was available for the hands-on tests. Published benchmarks placed the new models near leading competitors, but the review treated those rankings as preliminary and reserved a direct comparison until the full Grok 2 became available.

• Grok 2 Mini answered two snail-well reasoning problems correctly and produced a requested news summary of 125 words, squarely within the specified 100–150-word range. Its default writing was clear and well formatted, while Fun mode supplied a more playful tone without requiring a detailed style prompt from the user.

• The coding test required iteration. Grok 2 Mini first generated an attractive checkers interface without working game logic, then produced a functional version after a follow-up prompt, although the pieces became visually confusing. Real-time queries were less successful, returning incorrect Olympic context and outdated Nvidia stock information despite access to recent X posts.