Use Muse Wonder when you need an API, pay-as-you-go testing, or access on any Muse Code plan. Use Muse Wonder Full for demanding, interactive work inside Muse Code if you are willing to pay for Power Usage or Maximum Usage and your tasks benefit from its stronger published long-horizon coding and workflow scores. Setting Muse Wonder’s reasoning to its maximum does not convert it into Wonder Full; they are separate models.
Both models launched on October 12, 2026, and accept text, images, video, audio, and PDFs while returning text. The practical difference is access as much as capability: Wonder is the publicly priced API model, while Wonder Full is a Muse Code-only option with no documented API endpoint or per-token rate.
As an Amazon Associate I earn from qualifying purchases.
What Muse Wonder and Muse Wonder Full are
Meta AI Research introduced the family under the title “Muse Wonder: A Frontier Model For Personal & Professional AI Agents.” Wonder and Wonder Full are two distinct reasoning models, not two modes of one model. They are designed for multi-step agent work such as software engineering, repository maintenance, browser and desktop workflows, document analysis, and other professional tasks that require a sequence of decisions rather than a single answer.
The public model identifier for Wonder is muse-wonder-1.0. The corresponding Full identifier is muse-wonder-full-1.0, but that model is exposed through Muse Code rather than a documented public API. Both models are hosted proprietary services; no downloadable weights, parameter counts, or local-inference package are identified in the published comparisons.
#1 Best Overall
- Read Before You Buy — No Video Output: These adapters support charging and USB 2.0 data transfer, but cannot transmit video signals. Except for standard USB webcams (which use USB data only), they are not compatible with HDMI/DisplayPort cables, video-capable USB-C hubs, or docking stations with video output.
- Convert USB-A Ports to USB-C: Designed to connect USB-C earphones, cables, flash drives, card readers, and other USB-C accessories to standard USB-A ports. Plug-and-play with no drivers or software required.
- Aluminum Alloy Housing: Built with a sturdy aluminum alloy shell that aids in heat dissipation and protects against daily wear and scratches. Designed to maintain a stable and secure connection.
- Compact & Travel-Friendly: The ultra-compact design allows the adapter to stay plugged into your device without blocking adjacent ports or adding bulk, reducing wear and tear on your original USB ports.
- 12-Month Warranty: Backed by a 12-month manufacturer warranty for peace of mind. Designed to meet strict quality control standards for reliable everyday performance.
Shared input and output modalities
Each model accepts text, images, video, audio, and PDF files and returns text. That makes either model suitable for mixed-input tasks such as explaining a screen recording, extracting decisions from a meeting audio file, reviewing a PDF specification, or turning an image of an error into a troubleshooting plan. The modality list does not imply identical limits for every file type, so check the limits shown in the product you are using before designing a production workflow.
Context-window information
Muse Wonder has a documented context window of 1,048,576 tokens. Public comparison material does not state an equivalent exact limit for Wonder Full. Do not assume that Full has the same capacity, a larger one, or unlimited repository memory simply because it scores higher on some long-horizon benchmarks.
Access, plans, and price
Access is the first decision point. The following figures are the published 2026 terms and can change with preview status, geography, usage allowances, or plan revisions.
Recommended Free Tools
| Model | Where it is available | Published price | What the price means |
|---|---|---|---|
Muse Wonder (muse-wonder-1.0) |
Meta Model API and every Muse Code plan | $4 per million input tokens; $0.20 per million cached input tokens; $20 per million output tokens | Token billing for API use. The model is described as being in preview. |
Muse Wonder Full (muse-wonder-full-1.0) |
Muse Code only | Power Usage: $50/month; Maximum Usage: $200/month | Subscription access with usage allowances that reset. No public per-token API rate is listed. |
A Muse Code subscription credit is not API credit. Paying for Power Usage or Maximum Usage does not create a token balance that you can spend through the Meta Model API, and an API account does not unlock Wonder Full in Muse Code. Verify the current plan page, regional eligibility, allowance size, and preview terms immediately before purchase.
What API pricing looks like in practice
Wonder’s rates separate input, cached input, and output. At list price, one million uncached input tokens plus 100,000 output tokens would be $6: $4 for input and $2 for output. If the one million input tokens were billed as cached input instead, that input portion would be $0.20, making the same 100,000-token output example $2.20. These are arithmetic examples, not a promise about a minimum charge, rounding policy, rate limits, or future pricing.
Rank #2
- 5-in-1 USB-C Hub: Experience comprehensive connectivity featuring a Power Delivery input, two USB-A 2.0 ports, a USB-A 3.0 port, and an HDMI port. (Note: The USB-C power delivery input port is only for connecting an external wall charger to power your laptop and cannot power peripheral devices.)
- 90W Pass-Through Charging: Achieve optimal charging with 90W pass-through power to your laptop, supported by a total input of 100W, with the hub reserving 10W for operational efficiency. (Note: Wall charger not included.)
- Quick Data Transfers: Accelerate your productivity with rapid data transfers using a high-speed 5Gbps USB 3.0 port and two 480Mbps USB 2.0 ports.
- 4K HDMI Display: Enhance your visual experience with a hub capable of delivering 4K resolution at 30Hz in both mirror and extend modes. Please note that this hub is compatible with MacBook (macOS 12 and newer), Windows 10 and 11, ChromeOS, and laptops equipped with DP Alt Mode and Power Delivery. Note: This device is not compatible with Linux.
- What You Get: Anker USB-C Hub (5-in-1, 4K HDMI), welcome guide, 18-month warranty, and our friendly customer service.
Caching is most useful when an agent repeatedly sends a stable system prompt, tool schema, or codebase context. Measure your provider’s cache-hit behavior and freshness rules rather than assuming that every repeated token receives the cached rate.
How the published capability results differ
Available evaluations favor Wonder Full overall, but they do not establish a universal winner. LLM Stats reports Full winning 17 of 18 shared benchmarks and gives it an aggregate score of 53.1 versus 49.7 for Wonder. Those aggregates combine different tasks, so they are a directional signal rather than a guarantee for your prompts.
Free tools Windows power users keep installed
One-click scans. No signup required.
| Evaluation | Wonder Full | Muse Wonder or comparator | How to read it |
|---|---|---|---|
| DeepSWE v1.1 | 79.0 | Wonder 78.8 | Nearly tied in the reported Meta results; a small gap should not drive a purchase by itself. |
| SWE-Marathon | 57.5 | Wonder 53.1 | Full leads, but the test contains only 20 projects and uses different vendor harnesses. |
| AutomationBench-AA | 45.4 | GPT-6 Astra 41.6; Claude Opus 5.5 41.2 | Artificial Analysis reports Full ahead of the listed comparators on this automation test. |
| OSWorld V2.1 strict completion | 39.4% | Wonder value not stated | This is strict completion; partial-credit scores are not the same as fully completed workflows. |
| ProgramBench | 59.0 | Wonder 50.0; Claude Opus 5.5 70.5 | Full beats Wonder in the reported Meta results, while Opus 5.5 scores higher. |
| HealthBench Professional | 62.3 | Claude Opus 5.5 65.6; GPT-6 Astra 65.3 | Full trails both comparators; neither model should be treated as an unsupervised medical authority. |
| AA Intelligence Index v4.3 | 56 | Wonder 52; Claude Opus 5.5 58 | Full has a four-point lead over Wonder but remains below the listed Opus score. |
Meta’s figures and third-party evaluations use different harnesses, settings, and task definitions. The numbers are reported results, not independent hands-on testing in every environment. For a production decision, replay representative prompts and record success rate, tool-call errors, latency, human correction time, and total cost.
Choose by the job you need to do
Building an application or agent
Start with Muse Wonder. It is the family member with a documented Meta Model API, a stable model identifier, and transparent per-token rates. That combination lets you estimate a request budget, add retries and logging, and change traffic gradually. Wonder Full has no documented API endpoint in the available material, so it is not a suitable dependency for an application that must call a model programmatically.
Using Muse Code on an ordinary plan
Choose Wonder. It is available across Muse Code plans, whereas Full requires Power Usage or Maximum Usage. If your team already has a lower-tier subscription and only occasionally needs extended reasoning, upgrading solely for Full may cost more than the quality improvement saves.
Rank #3
- Sleek 7-in-1 USB-C Hub: Features an HDMI port, two USB-A 3.0 ports, and a USB-C data port, each providing 5Gbps transfer speeds. It also includes a USB-C PD input port for charging up to 100W and dual SD and TF card slots, all in a compact design.
- Flawless 4K@60Hz Video with HDMI: Delivers exceptional clarity and smoothness with its 4K@60Hz HDMI port, making it ideal for high-definition presentations and entertainment. (Note: Only the HDMI port supports video projection; the USB-C port is for data transfer only.)
- Double Up on Efficiency: The two USB-A 3.0 ports and a USB-C port support a fast 5Gbps data rate, significantly boosting your transfer speeds and improving productivity.
- Fast and Reliable 85W Charging: Offers high-capacity, speedy charging for laptops up to 85W, so you spend less time tethered to an outlet and more time being productive.
- What You Get: Anker USB-C Hub (7-in-1), welcome guide, 18-month warranty, and our friendly customer service.
Maintaining a large repository over many steps
Try Wonder Full inside Muse Code when the work is interactive and the higher subscription is acceptable. Full’s reported DeepSWE and SWE-Marathon results give it an edge on sustained software tasks. Still, repository size, test quality, tool permissions, and the agent harness can dominate the outcome. A model that scores well in a benchmark can fail if it cannot run your build, access the right branch, or recover from a broken migration.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problemsTerminal coding or greenfield programming
Do not assume Full wins every coding task. ProgramBench reports 59.0 for Full, 50.0 for Wonder, and 70.5 for Claude Opus 5.5; the published comparisons also place Opus 5.5 ahead on Terminal-Bench. If terminal execution and new-program construction are your main workloads, compare those systems on your own repository and shell environment rather than selecting from the aggregate score alone.
Health, legal, finance, or other high-consequence work
Treat both models as assistants that require qualified review. Full scores 62.3 on HealthBench Professional, below Claude Opus 5.5 at 65.6 and GPT-6 Astra at 65.3. A benchmark score does not validate a diagnosis, legal position, investment decision, or compliance conclusion. Use source documents, approval gates, audit logs, and a person accountable for the final decision.
Low-commitment experimentation
Use the Wonder API. Token billing lets you stop after a small experiment instead of taking on a $50 or $200 monthly Muse Code subscription. Track input and output tokens separately; long prompts and verbose agent traces can make an apparently cheap prototype expensive.
Using Muse Wonder through an API
The public information identifies the model and rates but does not publish a complete endpoint path, SDK command, authentication sequence, or rate-limit table. Do not copy an endpoint from an unrelated provider or invent one in production. Use the current Meta Model API documentation for the exact request format.
The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Rank #4
- Dual Converters, Infinite Potential：Includes 2× USB C male to USB A female adapters and 2× USB A male to USB C female adapters. Perfect for a wide range of uses—tablets with Bluetooth keyboards, expand USB ports on macbook, and more. Two different converters for all your daily needs
- Next-Level 10Gbps & 3A Charging: No more slow 480Mbps, this usb to usb c adapter has a transfer speed of up to 10Gbps, allowing you to do more transferring in less time. This usb adapter fits both USB A and USB C charger, supporting up to 3A fast charging
- Upgraded Exquisite Craftsmanship: With an aluminum alloy housing and metal connector, the usbc to usb adapter is extremely durable and sturdy. Rigorously tested to withstand more than 10,000 times of plugging and unplugging, ensuring long-lasting performance
- Broad Compatible: The usb c to usb adapter widely supports all USB C/ USB A devices like laptops, tablets, cellphones, car chargers, and phone chargers. Such as compatible with MacBook Pro/Air 2023/2022, Thunderbolt 4/3 Devices,Apple MagSafe Watch 9/8/7/SE/Ultra, iPad Pro 2022/2021, Samsung Galaxy S23/S20/S10, and iPhone 17/16/15 Pro. Plug and play
- Please Note: To reach 10Gbps speed, keep the cable under 3.3 ft. For USB A Male to USB C adapters, try flipping the USB C connector. USB C Male to USB A adapters support bidirectional 10Gbps transfer within 3.3 ft
- Confirm eligibility and preview terms. Check that your account and region can call
muse-wonder-1.0, and note any preview restrictions. - Create credentials using the provider’s current console. Keep the key server-side, store it in a secret manager, and never embed it in a browser bundle or mobile application.
- Send a minimal text request first. Record the model identifier, input tokens, output tokens, latency, HTTP status, and any request ID returned by the service.
- Add multimodal inputs deliberately. Test each image, video, audio, or PDF type your product will accept; validate file-size and duration limits before exposing the feature to customers.
- Implement operational controls. Set output limits, timeouts, retry rules, idempotency where supported, rate-limit backoff, and a human review path for consequential actions.
- Measure cost and quality on real tasks. Compare cached and uncached prompts, count tool-call failures, and sample outputs for factuality and instruction-following before increasing traffic.
There is no documented Full endpoint to substitute into that process. If an application requirement says “Full,” the requirement currently points to an interactive Muse Code workflow rather than a callable public service.
Using Wonder Full inside Muse Code
Wonder Full is the choice for a user who wants the strongest published Muse-family results while working inside the Muse Code environment. Power Usage costs $50 per month and Maximum Usage costs $200 per month in the cited 2026 terms. Both are subscription products with resetting allowances, not unlimited compute and not API balances.
When an upgrade is justified
- Your tasks regularly span many files, tests, and tool calls, and failures are expensive to diagnose.
- You can evaluate the result inside Muse Code rather than needing a standalone service endpoint.
- Your current plan does not provide enough allowance for the workload, and the expected reduction in rework exceeds the subscription cost.
- You have a rollback path when an automated edit changes behavior or damages data.
When to stay with Wonder
- You need predictable per-request accounting or customer-facing API access.
- Your work is mostly short prompts, code explanations, or small patches where Full’s benchmark advantage may not appear.
- You cannot commit to a $50 or $200 monthly plan.
- Your organization requires a documented API, region, retention policy, or integration contract that Muse Code does not provide.
Why the reasoning setting is not a substitute for Full
Muse Wonder may expose a reasoning control, including a maximum setting, but that setting changes how the Wonder model allocates effort. It does not load Wonder Full’s separate weights, training, or product access. If a workflow specifically requires Full, selecting maximum reasoning on Wonder is still using Wonder and should be tested and budgeted as Wonder.
This distinction matters for both quality and cost. A longer trace can increase output-token charges and latency without reproducing Full’s behavior on the benchmarks where Full leads. Conversely, Full’s higher subscription price is not proof that every prompt will improve; short deterministic tasks may show little difference.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Context, files, and long-horizon reliability
The one-million-token context documented for Wonder can accommodate substantial source material, but context length is not the same as reliable comprehension. Large prompts still need clear file boundaries, an explicit objective, acceptance tests, and instructions about which sources take precedence.
Best Value
- 5-in-1 Connectivity: Equipped with a 4K HDMI port, a 5 Gbps USB-C data port, two 5 Gbps USB-A ports, and a USB C 100W PD-IN port. Note: The USB C 100W PD-IN port supports only charging and does not support data transfer devices such as headphones or speakers.
- Powerful Pass-Through Charging: Supports up to 85W pass-through charging so you can power up your laptop while you use the hub. Note: Pass-through charging requires a charger (not included). Note: To achieve full power for iPad, we recommend using a 45W wall charger.
- Transfer Files in Seconds: Move files to and from your laptop at speeds of up to 5 Gbps via the USB-C and USB-A data ports. Note: The USB C 5Gbps Data port does not support video output.
- HD Display: Connect to the HDMI port to stream or mirror content to an external monitor in resolutions of up to 4K@30Hz. Note: The USB-C ports do not support video output.
- What You Get: Anker 332 USB-C Hub (5-in-1), welcome guide, our worry-free 18-month warranty, and friendly customer service.
For either model, long tasks are safer when you:
- Split discovery, planning, implementation, and verification into observable stages.
- Ask for a concise change plan before allowing writes.
- Run tests after each logical change instead of waiting until the end.
- Keep a clean branch or snapshot so an automated edit can be reverted.
- Require the agent to report failed commands rather than silently continuing.
Because Full’s exact context limit is not publicly stated in the available comparisons, avoid promising customers that it can ingest a particular repository size. Measure the largest prompt and file set that your Muse Code plan accepts.
Limitations and uncertainty to account for
- Availability can change. Both models launched in October 2026, and Wonder is described as being in preview. Prices, regional access, plan allowances, and model identifiers may change.
- Full’s API status is undocumented. The available information lists Muse Code access but no public endpoint or per-token price.
- Full’s context limit is not confirmed. Only Wonder’s 1,048,576-token window is explicitly documented.
- Benchmarks are not guarantees. Most figures come from Meta or named evaluators using their own harnesses and task definitions.
- No local fallback is identified. The models are hosted proprietary services, with no downloadable weights reported.
A practical selection checklist
- Need an API? Select Muse Wonder; Full has no documented public endpoint.
- Need Muse Code on an existing ordinary plan? Select Wonder, which is available across plans.
- Need long interactive repository work and can pay monthly? Pilot Wonder Full on Power Usage or Maximum Usage.
- Need the lowest commitment? Use Wonder’s token billing and cap spend at the application level.
- Work in a high-consequence domain? Choose either only with human review, provenance checks, and an approval process.
- Unsure whether the difference matters? Run a blind comparison on your own tasks and score completion, corrections, latency, and cost.
How to run a fair pilot
Create a small evaluation set that reflects the work you actually do: bug fixes with tests, multi-file refactors, document extraction, browser or desktop procedures, and failure recovery. Keep the prompt, files, tool permissions, and success criteria identical. For API trials, record token usage and cache hits. For Muse Code trials, record subscription allowance consumed and the number of manual interventions.
Score more than the final answer. A useful pilot records whether the model chose the right files, preserved existing behavior, handled a failed command, explained uncertainty, and stopped when approval was required. Include at least one task where a plausible but unsafe answer would be costly. The result should tell you whether Full’s higher subscription changes your completed-work rate, not merely whether its benchmark reputation sounds better.
Bottom line
Muse Wonder is the default for developers: it is API-accessible, available on every Muse Code plan, and priced by token. Muse Wonder Full is the specialist upgrade for demanding Muse Code sessions where the reported long-horizon and automation advantages justify Power Usage or Maximum Usage. Neither choice removes the need to test on your own workload, and maximum reasoning on Wonder is not a hidden route to Full.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
