Skip to content

Grok 4.7 · Writing

Grok 4.7 for writing: our test runs

Grok 4.7 ran our 5 writing prompts in llmwise, through the same pipeline your messages take. Here's every reply as it came, whether it passed, what it cost and what it counts as on Pro, next to the other models.

Based on 5 of our test runs on , through OpenRouter with the app's own prompt and settings.

Short answer

In our test runs on September 27, 2026, Grok 4.7 passed 3 of 5 of our writing prompts (both hard ones): tied 11th of 15 models. Each reply counted as 1 message on Pro, where it has 250 a month, and cost $0.0042 to run on average.

Prompt by prompt

Each of the 5 writing prompts, and how Grok 4.7's reply did.

Grok 4.7 on each writing prompt
PromptResultCostTime
Announce a second bakery shop on LinkedInEverydayFailed: Graded 4.5 of 5 on average (lowest 4); but 94 words, under the 100 asked for.$0.00224.0 s
Rewrite corporate jargon in plain wordsEverydayFailed: Graded 3.7 of 5 on average (lowest 3).$0.008819.3 s
Decline a meeting and offer two timesEverydayPassed: Graded 4.7 of 5 on average (lowest 4).$0.00192.3 s
A product announcement with five rulesHardPassed: Graded 4.0 of 5 on average (lowest 3).$0.00172.6 s
Argue both sides of free busesHardPassed: Graded 4.3 of 5 on average (lowest 4).$0.006512.1 s

Every writing reply from Grok 4.7

  1. Announce a second bakery shop on LinkedIn

    An everyday prompt: “Announce a second bakery shop on LinkedIn”, in full.

    Failed: Graded 4.5 of 5 on average (lowest 4); but 94 words, under the 100 asked for.

    We're opening a second Crumb & Co. shop in Leeds on 14 November.
    
    The new bakery will offer the same bread, cakes and pastries you already know from our first shop, baked fresh each morning. If you've been asking for a Leeds location, this one is for you.
    
    On opening day, coffee is free with any purchase. Come in, say hello, and try something from the counter. We'd love to meet you and hear what you'd like to see on the shelves.
    
    Thank you for supporting a small bakery. We can't wait to welcome you.

    1,624 tokens in, 207 out (93 of them reasoning) · 4.0 s · $0.0022 · 1 message on Pro · answered by x-ai/grok-4.7 via xAI ·

  2. Rewrite corporate jargon in plain words

    An everyday prompt: “Rewrite corporate jargon in plain words”, in full.

    Failed: Graded 3.7 of 5 on average (lowest 3).

    In the third quarter, teams from different departments worked together on an initiative that put a stronger customer focus into practice. Retention measures rose 12%, and the speed at which enterprise customers left fell by 30 basis points. We also resized customer contact points across channels so they used available capacity more efficiently.

    1,657 tokens in, 1,631 out (1,571 of them reasoning) · 19.3 s · $0.0088 · 1 message on Pro · answered by x-ai/grok-4.7 via xAI ·

  3. Decline a meeting and offer two times

    An everyday prompt: “Decline a meeting and offer two times”, in full.

    Passed: Graded 4.7 of 5 on average (lowest 4).

    Subject: Re: Vendor demo Wednesday
    
    Hi Tom,
    
    Thanks for the invitation. I can't make the vendor demo on Wednesday at 11:00 — I'll be presenting to the board then.
    
    I'm free Thursday at 10:00 or Friday at 14:00 if either works. Happy to join then.
    
    Best regards

    1,629 tokens in, 138 out (74 of them reasoning) · 2.3 s · $0.0019 · 1 message on Pro · answered by x-ai/grok-4.7 via xAI ·

  4. A product announcement with five rules

    A hard prompt: “A product announcement with five rules”, in full.

    Passed: Graded 4.0 of 5 on average (lowest 3).

    The Nordvik 2 desk lamp dims in five steps and charges from USB-C. Its arm turns 270 degrees, so you can aim the light where you need it. It costs €79 and ships on 1 December. Ready to light your desk?

    1,641 tokens in, 151 out (98 of them reasoning) · 2.6 s · $0.0017 · 1 message on Pro · answered by x-ai/grok-4.7 via xAI ·

  5. Argue both sides of free buses

    A hard prompt: “Argue both sides of free buses”, in full.

    Passed: Graded 4.3 of 5 on average (lowest 4).

    Making buses free would remove the fare barrier that keeps low-income riders from jobs, clinics, and schools, and it would cut the cost and delay of collecting fares. More people would leave cars behind, easing congestion and pollution while filling seats that already run half empty. A city that treats transit as a public service, like streets or parks, can fund it through broad taxes and gain mobility, equity, and cleaner air in return.
    
    Free buses sound generous, but they shift a real cost onto taxpayers who may never ride, while inviting overcrowding, slower trips, and service that cannot keep up with demand. Without fares, agencies lose a direct signal of what routes people value and a revenue stream that pays drivers and buys buses. Cities that have tried fare-free service often face vandalism, disorder, and cuts elsewhere. Mobility is better served by targeted discounts for those who need them and reliable, frequent service that people will pay to use.

    1,618 tokens in, 1,159 out (968 of them reasoning) · 12.1 s · $0.0065 · 1 message on Pro · answered by x-ai/grok-4.7 via xAI ·

How it compares on writing

Grok 4.7 beside the job's picks among every model. 2 models share the hard-problems pick, and Grok 4.7 isn't one of them.

Grok 4.7 beside other models on writing
ModelPassedOn ProCost per reply
GLM 5.3Our pick: best value4 of 5250 a month on Pro$0.0006
GPT-6 LunaOur pick: hard problems (shared), everyday5 of 560 a day on Pro$0.0001
Grok 4.73 of 5250 a month on Pro$0.0042
On Pro: Pro's count on each model. Cost: what OpenRouter charged us per reply, on average.

Every limit is published. Paid plans also have a monthly fair-use limit on AI cost: Pro $7.50, Max $20, Ultra $42, Studio $85. Using every message on your plan at typical sizes stays under it; very large messages and heavy research use it faster. Every limit, explained.

How these runs were done

Rubric (graded). The grader model scores the reply from 1 to 5 on each published criterion. It passes with an average of 4 or more and no criterion under 3, and only if it also meets the prompt's automatic rules (length, words it must or mustn't use).

The grader is Claude Opus 5.5 at low reasoning effort; its own replies are graded by GPT-6 Astra, so no model grades itself. Its prompt and every rubric are on the methods page.

How the runs were done, and every writing prompt.

Grok 4.7, and writing, elsewhere

Questions

Is Grok 4.7 good for writing?

In our test runs it passed 3 of 5 writing prompts, tied 11th of the 15 models in llmwise. Every reply is on this page, so you can judge them yourself.

How many of my messages does a writing reply from Grok 4.7 use?

1 message each on Pro, where it has 250 a month on Pro. The price of a message is fixed and shown before you send it, however long the reply.

How were these runs done?

The same way for every model: each prompt sent through llmwise's own pipeline, each reply checked the same way. The methods page has every prompt and how each is scored.

Claude, GPT, Gemini, DeepSeek, Grok, Kimi, and GLM, in one chat.

See what a message costs before you send it. Free is 5 messages to try; sign in with an email link, no password or card.