NEW · Morning journal prompts → start your day with intention
Random Prompts
Muse Wonder 1.0 — Meta AI — launched October 12 · 1M context · text + image + video + audio input

Meta Muse Wonder Prompt Generator

The Meta Muse Wonder prompt generator gives you 20 free, copy-paste prompts for Meta's new frontier multimodal AI — released October 12, 2026. Muse Wonder accepts text, images, video, and audio, with a 1-million-token context window for reading entire codebases, document sets, and meeting recordings in one session. No signup needed.

What is the Meta Muse Wonder Prompt Generator?

The Meta Muse Wonder prompt generator on this page provides 20 free, professionally crafted prompts for Muse Wonder 1.0, the frontier multimodal AI model released by Meta AI on October 12, 2026. Muse Wonder is Meta's most powerful model to date and a direct competitor to ChatGPT, Claude, and Gemini — built for long-context reasoning, multimodal analysis, and professional tasks that require reading and synthesising large inputs.

Unlike Muse Spark (which specialises in generating interactive web apps), Muse Wonder is optimised for understanding: reading an entire codebase, a set of legal contracts, or hours of meeting transcripts in a single 1-million-token session. It accepts text, images, video clips, and audio files natively — so you can attach a screenshot and a PDF and a voice memo in one prompt, and Muse Wonder will reason across all three.

The 20 prompts below cover Muse Wonder's two defining use cases: multimodal analysis (inputs combining images, video, or audio with text) and long-context reasoning (reading multiple long documents, codebases, or transcripts in one session). Each prompt is structured to take advantage of Muse Wonder's specificity preference — detailed output instructions produce significantly better results than open-ended requests.

How to Prompt Meta Muse Wonder: Two Modes

Multimodal Analysis Mode

Analyze [this image/video/audio]. Identify: [specific things to look for]. Format the output as [table/bullets/document]. Write for [audience].
  • Name the input type explicitly (screenshot, video, audio)
  • Give a specific list of what to look for
  • Specify the output format and audience
  • Attach all files in a single prompt
  • Describe each attached file briefly

Long-Context Reasoning Mode

Read [these N documents]. Find: [specific question]. Format: [output structure]. Audience: [who will read this]. Length: [word count or page count].
  • State the overall goal before attaching documents
  • List exact output sections you need
  • Give word count or page constraints
  • Specify the audience and their knowledge level
  • Ask it to flag what it couldn't find

20 Free Meta Muse Wonder Prompts — Copy & Paste

Click any prompt to copy — paste into Meta AI with your attached files

1. Analyze a Screenshot for UX Issues

Multimodal

Analyze this screenshot of my web app's checkout flow. Identify: (1) any UX friction points — elements that slow users down or cause confusion; (2) accessibility issues — contrast, target size, label clarity; (3) trust signals that are present or missing; (4) the single highest-impact improvement I could make in under 2 hours of dev time. For each issue, rate severity: Critical / Should Fix / Nice to Have. [Attach screenshot]

2. Research Synthesis From Multiple Long Sources

Long-Context

Read these three research papers and write a 600-word synthesis that: (1) identifies the single key finding each paper contributes; (2) shows where the three papers agree; (3) shows where they contradict each other and why — methodology difference, sample size, or definitional disagreement; (4) identifies the most important question the combined research leaves unanswered. Write for a policy audience: confident, direct, no academic hedging. [Attach papers]

3. Legal Contract Risk Review

Professional

Review this contract for risk to the party named [Your Company Name]. Identify: (1) clauses that expose us to unlimited liability; (2) IP ownership provisions and what we are giving up; (3) termination conditions we cannot control; (4) payment terms and what happens if we miss one; (5) any clause that is unusual compared to standard industry practice and why it matters. For each issue give: the clause reference, the risk in plain language, and the specific change I should request. [Attach contract PDF]

4. Describe a Product Video for Social Copy

Multimodal

Watch this short product demo video and write: (1) a 60-second script for a voice-over that highlights the three strongest value propositions shown; (2) three 280-character tweets that could accompany the video on launch day — each using a different angle (feature, outcome, social proof); (3) a 150-word LinkedIn post for the CEO to share on launch day. Keep all copy benefit-led, not feature-led. Avoid superlatives. [Attach video file]

5. Entire Codebase Audit for Security

Long-Context

Review the attached codebase for security vulnerabilities. Focus on: authentication and session handling; SQL/NoSQL injection vectors; hardcoded secrets or credentials; unsafe deserialization; CORS configuration; third-party dependency risks in package.json/requirements.txt. For each vulnerability found: severity (Critical / High / Medium / Low), the file and line, how it could be exploited in a realistic attack scenario, and the exact code change needed to fix it. List what you checked and found clean, too. [Attach codebase]

6. Meeting Transcript to Decisions + Action Items

Long-Context

Read this full meeting transcript and extract: (1) every decision made — word it as a statement of what was decided, not who said what; (2) every action item with owner and deadline as stated; (3) every commitment made to a person not in the room; (4) any open question that was raised but not resolved. Format as a structured document I can email to all attendees. If an owner or date was not stated, write 'unassigned' or 'no date set'. [Paste transcript]

7. Audio Briefing Transcription + Summary

Multimodal

Transcribe this audio recording and then produce: (1) a clean, formatted transcript with speaker labels if multiple speakers are present; (2) a 150-word executive summary of the main points; (3) a list of any names, numbers, dates, or proper nouns mentioned that I should double-check for accuracy; (4) the single most important thing I need to act on from this recording. [Attach audio file]

8. Competitive Intelligence From Annual Reports

Long-Context

Read these three annual reports from my main competitors. For each one, identify: their stated strategic priorities for the coming year; any new product lines, markets, or partnerships announced; any acknowledged risks that are relevant to us; and changes in R&D spend. Then write a 400-word briefing for our leadership team comparing their trajectories and flagging where we are most exposed. [Attach PDFs]

9. Image-Based Data Extraction

Multimodal

Extract all the data from this chart image into a clean CSV. Then: (1) describe the trend in two sentences as if presenting to a board; (2) identify the highest and lowest data points with their labels; (3) flag any anomaly or break in trend that would warrant further investigation; (4) state the single most important business implication of this data. [Attach chart image]

10. Technical Documentation Review

Long-Context

Review this technical documentation against the actual codebase I have attached. Identify: (1) anywhere the documentation contradicts the code; (2) public API endpoints or parameters that are in the code but not documented; (3) documented features that no longer exist in the code; (4) examples in the documentation that are syntactically incorrect. Format as a prioritised list: Documentation Errors / Missing Coverage / Outdated Examples. [Attach docs and code]

11. Multimodal Product Feedback Analysis

Multimodal

I am attaching screenshots of 50 user reviews from our app store listing. Read each one and produce: (1) the top 5 most frequently mentioned complaints, ranked by frequency; (2) the top 5 most frequently mentioned praise points; (3) three distinct user segments visible in the feedback — describe each by what they value and what frustrates them; (4) the one product change that would address the most complaints simultaneously. Use exact quotes to support each finding. [Attach screenshots]

12. Medical Research Summary for Patient

Long-Context

Read this clinical study and summarize it for a patient with no medical background. Include: (1) what the study was testing in one plain-English sentence; (2) what they found — the main result in terms a non-specialist can understand; (3) what the limitations are — small sample, short follow-up, specific population; (4) what this means for someone with this condition in practical terms. Do not use statistical jargon. Do not overstate certainty. If the study is preliminary, say so clearly. [Attach PDF]

13. Video-to-Tutorial Outline

Multimodal

Watch this tutorial video and produce a written step-by-step guide that someone could follow without watching the video. Include: timestamps for each major section, exact commands or actions shown, expected outputs after each step, and any caveats mentioned verbally. Format as a numbered list with clear section headers. Note any step where the video moves too fast or skips an assumption that a beginner would need explained. [Attach video]

14. Financial Statement Analysis

Long-Context

Read the attached quarterly financial statements for the last four quarters. Calculate and explain: (1) revenue growth rate quarter-over-quarter; (2) gross margin trend; (3) operating expense as a percentage of revenue — is it improving or worsening; (4) cash runway if current burn rate continues. Then write a 300-word narrative summary for a non-financial founder that explains the overall trajectory honestly — good and bad. [Attach financial statements]

15. Podcast Transcript to Blog Post

Long-Context

Read this podcast transcript and turn it into a 1,200-word blog post. Keep the speaker's opinions, examples, and specific data points — these are the most valuable parts. Structure: a compelling opening that hooks readers who did not listen to the podcast; 3–4 sections with descriptive H2s covering the main ideas; a closing paragraph with one concrete takeaway. Remove filler phrases and repetition. Do not say 'In this podcast' or 'The host asked' — write as if the ideas are being shared directly. [Paste transcript]

16. Competitor Website Teardown

Multimodal

I am attaching screenshots of my main competitor's website — homepage, pricing page, and features page. Analyze: (1) how they position their product and who their primary buyer persona appears to be; (2) their key value propositions and how they communicate them; (3) their pricing strategy and what signal it sends; (4) any UX patterns that are working particularly well; (5) three things we should copy and one thing they are doing wrong. Be specific — reference elements visible in the screenshots. [Attach screenshots]

17. Entire Book Chapter Critique

Long-Context

Read this draft chapter and give me a professional editor's critique. Structure your feedback as: (1) Argument — is the central claim clear and supported throughout? Where does it weaken? (2) Structure — does the chapter flow logically? Where does a reader lose the thread? (3) Prose — identify the three weakest sentences or paragraphs and rewrite each; (4) Evidence — are the claims supported? Where does the author assert without proof? (5) Priority fix — the single change that would most improve this chapter. [Paste chapter]

18. Image to Accessibility Audit

Multimodal

Examine this screenshot of my website and evaluate it against WCAG 2.2 AA accessibility standards. For each issue found, provide: the WCAG criterion it violates (e.g. 1.4.3 Contrast), the specific element on screen, the problem in plain language, and the change needed to pass. Then rate the overall accessibility: what is working well; what is the single most critical issue to fix first; and the estimated effort level (High / Medium / Low) for each fix. [Attach screenshot]

19. Write a Feasibility Study

Professional

Write a 2-page feasibility study for this project: [describe the project — e.g. 'launching a B2B SaaS product for property managers in the UK']. Structure: (1) Executive Summary — is this feasible and on what conditions; (2) Market Analysis — size, growth, key players, our position; (3) Technical Feasibility — what needs to be built and key risks; (4) Financial Feasibility — estimated costs, revenue assumptions, break-even timeline; (5) Recommendation — go, no-go, or go with conditions. Write for a board audience: confident, evidence-based, no filler.

20. Cross-Document Consistency Check

Long-Context

I am attaching our product specification document and our marketing website copy. Read both documents completely and identify: (1) any claim in the marketing copy that the spec document does not support; (2) features listed in the spec that are absent from the marketing copy; (3) any terminology that is used inconsistently between the two documents; (4) pricing or limitations mentioned in one document but not the other. Format as a table: Issue / Location in Document A / Location in Document B / Recommended Correction. [Attach both documents]

Muse Wonder vs. ChatGPT vs. Claude vs. Gemini (October 2026)

Muse Wonder's multimodal input breadth and long context window are its key differentiators:

Model Context Window Input Modalities API Price (input)
Muse Wonder 1.0 (Meta) ★ 1M tokens Text + Image + Video + Audio $4 / 1M tokens
GPT-5.5 (OpenAI) 128K tokens Text + Image $2 / 1M tokens
Claude Sonnet 5.5 (Anthropic) 1M tokens Text + Image $2 / 1M tokens
Gemini 4 (Google) 2M tokens Text + Image + Video + Audio $6 / 1M tokens
Muse Spark v1.3 (Meta) 200K tokens Text + Image + Voice Free (consumer apps)

★ Muse Wonder 1.0 released October 12, 2026 by Meta AI. Muse Wonder Full (stronger variant) announced but not yet publicly available. Pricing approximate at launch.

Meta Muse Wonder Prompting Tips

Do:

  • State your goal before attaching documents — context first, task second
  • Specify exact output format: table, numbered list, 300-word document
  • Name the audience — "a board", "a patient", "a junior developer"
  • For multimodal: attach all files at once and describe each briefly
  • Ask it to flag gaps — "note anything you could not verify from the documents"
  • Request severity ratings for issues: Critical / High / Medium / Low

Avoid:

  • Vague open-ended questions — "what do you think about this?" produces poor results
  • Attaching files without describing what they are
  • Asking for analysis without specifying the output structure
  • Omitting the audience — the same content needs very different writing for a board vs. a technician
  • Expecting Muse Wonder to render apps — that's Muse Spark's capability, not Muse Wonder's
  • Using it for quick one-line tasks — its long-context strength is wasted on short inputs

Frequently Asked Questions — Meta Muse Wonder

What is the Meta Muse Wonder prompt generator?

The Meta Muse Wonder prompt generator on this page provides 20 free, copy-paste prompts for Meta's Muse Wonder 1.0, released October 12, 2026. Muse Wonder is Meta's new frontier multimodal AI model — it accepts text, image, video, and audio inputs and produces text output. Its standout capabilities are a 1-million-token context window (for reading entire codebases, books, and document sets in a single session) and native multimodal reasoning that combines analysis across text, images, video frames, and audio in one response. Available via API at $4 per million input tokens.

What is Meta Muse Wonder and who made it?

Muse Wonder 1.0 is a frontier multimodal large language model released by Meta AI (Meta Superintelligence Labs) on October 12, 2026. It is Meta's most capable AI model to date and is positioned as a direct competitor to ChatGPT, Claude, and Gemini. Unlike Muse Spark (which specialises in visual coding and rendered web apps), Muse Wonder is optimised for long-context reasoning, multimodal analysis, and professional-grade tasks requiring reading and synthesising large amounts of information. A stronger variant, Muse Wonder Full, was announced simultaneously but is not yet available via public API.

What makes Muse Wonder different from Muse Spark?

Muse Spark and Muse Wonder are different products with different strengths. Muse Spark specialises in visual coding — generating functional interactive web apps and dashboards rendered directly in Meta AI — and writing. Muse Wonder is optimised for reasoning over large inputs: it accepts text, images, video clips, and audio files, and can process up to 1 million tokens in a single session (roughly 750,000 words, or an entire large novel plus supporting documents). Muse Wonder excels at legal analysis, codebase review, research synthesis, financial modelling, and any task that requires reading many long documents at once.

What makes Muse Wonder different from ChatGPT and Claude?

Muse Wonder's key differentiators are: (1) native audio and video input — it can analyse a product demo video or meeting recording directly without conversion; (2) a 1M token context window, larger than most competitors' default context; (3) Meta's distribution reach — Muse Wonder is accessible through Meta AI's existing platforms alongside its API. Compared to GPT-5.5 and Claude Sonnet 5.5, Muse Wonder's multimodal input breadth (adding native audio and video to the standard text+image) is the primary differentiator. The 'Full' variant, when released, is expected to close the gap on reasoning tasks where Claude Opus and GPT-5.5 Pro currently lead.

How do I write a good Meta Muse Wonder prompt?

Muse Wonder responds best to prompts that are specific about: (1) the input you are providing and its format (PDF, screenshot, audio file, video clip); (2) the exact output structure you need — bullet lists, tables, a 300-word summary, or a specific document format; (3) the audience for the output — a board, a patient, a developer, a junior team member; (4) what to include and what to omit. For long-context tasks, describe the overall goal first, then attach the documents. For multimodal tasks, attach all files at once and describe each one briefly before giving the task.

Where can I access Meta Muse Wonder?

As of its October 12, 2026 launch, Meta Muse Wonder 1.0 is available via Meta's API at $4 per million input tokens. API access is through the Meta for Business developer portal. It is expected to be rolled out to Meta AI consumer products (meta.ai, WhatsApp, Instagram, Facebook) in the weeks following launch, starting with Meta AI Pro subscribers. The more powerful Muse Wonder Full variant was announced alongside the launch but does not yet have a public release date or API pricing.

More AI Model Prompt Generators

Related prompt generators

More free generators in this collection — no signup, unlimited use.

🔷Mistral Large 4 Prompt Generator

20 copy-paste prompts for Mistral Large 4 — cybersecurity, finance, engineering and multimodal analysis.

⚡Claude Haiku 5.5 Prompt Generator

20 copy-paste prompts for Claude Haiku 5.5 — high-volume classification, summarization and batch tasks at $0.10/M tokens.

🎼Claude Sonnet 5.5 Prompt Generator

20 copy-paste prompts for Claude Sonnet 5.5 by effort level — fast agentic coding and visual tasks.

🏛️Claude Opus 5.5 Prompt Generator

20 free prompts for Claude Opus 5.5 — deep reasoning for long documents and audits.

☀️GPT-6 Sol Prompt Generator

20 copy-paste prompts for OpenAI GPT-6 Sol — coding, agents and long documents.

🌙GPT-6 Luna Prompt Generator

20 free prompts for OpenAI GPT-6 Luna — fast, affordable everyday tasks.

All AI text model prompt generators →