🔍 Read the full analysis: Muse Wonder Full Joins The AI Frontier, And Muse Wonder Charges Too on ThorstenMeyerAI.com
Get movie nights delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
Artificial Analysis rated Meta’s Muse Wonder Full 56 on its Intelligence Index and the API-available Muse Wonder 52. The independent evaluation found major gains over Muse Spark 1.3, but estimated a Muse Wonder index-task cost of $4.52—well above some similarly scored rivals. Muse Wonder Full is available only through Muse Code, with no published per-token price.
Meta’s new Muse Wonder Full scored 56 on Artificial Analysis’ Intelligence Index, while the API model Muse Wonder scored 52, according to an evaluation published October 12. The results show substantial gains over Muse Spark 1.3, but the analysis estimated that one index task on Muse Wonder costs $4.52—more than several competing models with similar scores.
Artificial Analysis said Muse Wonder Full improved on all 10 evaluations in its Intelligence Index v4.3 and placed in the top three on six. Its score of 56 is eight points above Muse Spark 1.3’s 48. Among 27 configurations compared in the source report, Claude Opus 5.5 at its maximum setting scored 58, while Claude Sonnet 5.5 at maximum also scored 56. These are benchmark results, not a guarantee of performance on every task.
The largest reported change was on Terminal-Bench 4.0, an evaluation of whether a model can complete work in a shell: Muse Wonder Full scored 57%, up from 33% for Muse Spark 1.3. It also scored 1,857 Elo on GDPval-AA, which measures economically useful knowledge work. The report placed that result between Claude Opus 5.5 at 1,866 and Claude Sonnet 5.5 at 1,838, while noting the results fall within the models’ confidence intervals.
The API-available Muse Wonder scored 52, the same Intelligence Index score reported for GPT-6.1 Sol. Artificial Analysis estimated a cost of $4.52 per index task for Muse Wonder, compared with $0.72 for Sol. The report’s task-cost figures are estimates based on the evaluation and model pricing, not a universal price for every user workload.
Infographic 01 · Where Meta’s new pair lands
Muse Wonder Full ties for second. Muse Wonder ties GPT-6.1 Sol.
Artificial Analysis scores 27 frontier and open-weight configurations on ten evaluations. Meta’s two new models land near the top of the list. Six weeks earlier, Muse Spark 1.3 scored 48.
Intelligence Index score, 27 configurations
Higher is better. Reasoning effort in brackets, as Artificial Analysis tested it.
Availability: Muse Wonder runs on Meta’s API and in Muse Code. Muse Wonder Full is only available to Muse Code users and has no per-token price.
Source: Artificial Analysis, Muse Wonder analysis (12 Oct 2026) and model leaderboard, read 13 Oct 2026. Intelligence Index v4.3, ten evaluations.
Powered by Thorsten Meyer AI
Strong Scores, Higher API Costs
The release gives developers and businesses two different decisions rather than one straightforward model choice. Muse Wonder is available through Meta’s API and in Muse Code, but its reported task cost is markedly higher than that of models with similar index scores. Muse Wonder Full posts stronger results but is restricted to Muse Code and has no per-token price in the source material, making its cost harder to compare.
Artificial Analysis’ figures suggest that benchmark capability alone may not determine the practical value of the API model. At the reported prices, a team running many evaluation-like tasks could face substantially different bills depending on model choice. Actual costs will vary with prompts, output length, caching and the work being done, so the index-task estimate should be treated as a comparison under the report’s test conditions rather than a forecast for every deployment.
The evaluation also records progress in coding and automation, alongside remaining reliability concerns. Muse Wonder Full’s reported hallucination rate was 23%, down from 33% for Muse Spark 1.3; Muse Wonder’s was 27%. Those figures indicate improvement in the tested measure, but they do not remove the need to verify outputs where errors could carry financial or operational consequences.
As an affiliate, we earn on qualifying purchases.
Two Models, Different Access
Meta shipped Muse Wonder and Muse Wonder Full six weeks after Muse Spark 1.3, according to the source report. Both are proprietary models with a reported one-million-token context window. They accept text, images, speech or audio, and video as input, and return text. Muse Wonder can be accessed through Meta’s API and within Muse Code, the company’s coding agent. Muse Wonder Full is available only to Muse Code users.
The listed Muse Wonder API rates are $4 per million input tokens, $0.20 per million cached input tokens and $20 per million output tokens. The report says those rates match Claude Opus 5.5’s list prices. Muse Wonder Full has no per-token price stated in the source, so its cost per task cannot be calculated from those rates. The source’s author also reports that Muse Spark 1.3 was priced at $1.25 per million input tokens and $4.25 per million output tokens.
Artificial Analysis’ Coding Agents Index compares model-and-harness combinations rather than models alone. It gave Muse Code with Muse Wonder Full a score of 68, equal to Claude Code with Claude Sonnet 5.5; Muse Code with Muse Wonder scored 63. The report cautions that this does not establish how Muse Wonder Full would perform inside Claude Code. It measures each model paired with its own coding environment.
“The difference is the bill: one Intelligence Index task costs $4.52 on Muse Wonder and $0.72 on Sol.”
— Thorsten Meyer, ThorstenMeyerAI.com
AI model performance benchmarking tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Costs Beyond the Benchmark
The available information does not state how much Muse Wonder Full costs per task or provide a per-token price for it. It is also unclear whether Meta will make that model available through its API or publish pricing. Without those details, buyers cannot make a like-for-like cost comparison between Full and the API model using the reported figures.
The $4.52 estimate applies to one Artificial Analysis Intelligence Index task under the evaluation’s workload and pricing assumptions. It should not be read as a fixed charge for each user request. The source excerpt also cuts off during its breakdown of Muse Wonder’s input-token costs, so it does not provide the complete accounting detail behind the estimate. The evaluation does not establish how the models will perform across every real-world task, or whether the reported gains will persist in independently tested deployments.
As an affiliate, we earn on qualifying purchases.
Pricing and Access Questions
For customers weighing the models, the next practical steps are to compare them on their own tasks and monitor Meta’s access and pricing announcements. Muse Wonder API users can measure token consumption against their expected workloads, while Muse Code users can test whether Wonder Full’s benchmark strengths translate to their coding needs.
The source material does not identify a further release date, planned API expansion for Muse Wonder Full, or a change to the listed rates. Until Meta provides those details, the cost comparison remains tied to the current published pricing and Artificial Analysis’ evaluation. Readers should also watch for further independent testing, particularly on task-level reliability and costs outside the benchmark setup.
As an affiliate, we earn on qualifying purchases.
Key Questions
How did Muse Wonder Full score?
Artificial Analysis gave Muse Wonder Full a score of 56 on its Intelligence Index v4.3. The report says it improved on all 10 evaluations and placed in the top three on six.
What is the difference between Muse Wonder and Muse Wonder Full?
Muse Wonder scored 52 and is available through Meta’s API and Muse Code. Muse Wonder Full scored 56 but, according to the report, is available only through Muse Code and has no stated per-token price.
How much does Muse Wonder cost?
The listed API prices are $4 per million input tokens, $0.20 per million cached input tokens and $20 per million output tokens. Artificial Analysis estimated $4.52 for one Intelligence Index task; actual costs depend on usage and task.
Is Muse Wonder Full better than Claude Opus 5.5?
The reported Intelligence Index score was 56 for Muse Wonder Full and 58 for Claude Opus 5.5 at its maximum setting. The source also reports Muse Code with Wonder Full tied with Claude Code using Sonnet 5.5 on the Coding Agents Index, which compares model-and-harness combinations—not the models in identical environments.
Are the new models free of hallucinations?
No. The source reports hallucination rates of 23% for Muse Wonder Full and 27% for Muse Wonder in the tested measure. Those results show improvement over Muse Spark 1.3, but do not establish that either model is reliable on every task.
Source: ThorstenMeyerAI.com
Fall Picks
fall essentials
As an affiliate, we earn on qualifying purchases.
