🔍 Read the full analysis: Muse Wonder Full Steps Into AI’s Frontier As Muse Wonder Charges on ThorstenMeyerAI.com
Prime made for students and young adults
- Fast, free delivery for dorm and study essentials
- Prime Video and Amazon Music included
- Member-only deals
TL;DR
Meta has released Muse Wonder and Muse Wonder Full, two proprietary models that Artificial Analysis evaluated on October 12, 2026. Muse Wonder Full scored 56 on the Intelligence Index, while the API-accessible Muse Wonder scored 52; Artificial Analysis estimated a benchmark task with Muse Wonder costs $4.52, compared with $0.72 for GPT-6.1 Sol at the same score.
Meta has released Muse Wonder and Muse Wonder Full, two proprietary AI models that Artificial Analysis evaluated on October 12, 2026. The stronger model scored 56 on the Intelligence Index v4.3 but is limited to Meta’s Muse Code product, while the API model scored 52 and carries an estimated $4.52 cost per benchmark task.
Artificial Analysis’ evaluation places Muse Wonder Full eight points above Muse Spark 1.3, which scored 48. The evaluator said Full improved on all ten tests in the index and ranked in the top three on six. Its score of 56 was below Claude Opus 5.5 at 58 and tied with Claude Sonnet 5.5 at 56 among the 27 configurations cited in the source. These are benchmark results, not a guarantee of performance on every real-world task.
The models share a one-million-token context window and accept text, images, audio or speech, and video as input, returning text. Meta makes Muse Wonder available through its API and inside Muse Code. Its list price is $4 per million input tokens and $20 per million output tokens, with cached input priced at $0.20 per million tokens. Muse Wonder Full is available only to Muse Code users, and Meta does not provide a per-token price for it.
Artificial Analysis estimated that a single Intelligence Index task costs $4.52 with Muse Wonder. The report puts GPT-6.1 Sol at $0.72 for the same task and score of 52, Gemini 4 Argon at $1.99 for a score of 53, and Claude Sonnet 5.5 at $5.46 for a score of 56. These are evaluator calculations based on the test’s token usage and the models’ prices, rather than a quoted price for a standard user request.
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 Meet High API Costs
The results give developers and businesses a concrete trade-off to weigh: benchmark capability versus estimated operating cost. Muse Wonder Full’s results suggest Meta has moved closer to leading models on several evaluated tasks, while its API model’s score is accompanied by a higher estimated task bill than some competitors with similar index results.
Artificial Analysis attributes much of Muse Wonder’s estimated expense to its input-token use. It says the model processed 7.69 million input tokens per index task, compared with 6.03 million for Muse Spark 1.3. The report estimates that input accounted for $3.43, or 76%, of the $4.52 task cost. Muse Wonder Full processed 6.44 million input tokens, but its per-token price is unavailable, so the same cost comparison cannot be made for that model.
For organizations considering an API, the distinction between a benchmark score and the full cost of running their own workloads matters. The report’s results can inform model selection, but actual expenses depend on how a service is used, its prompts, output length, and provider billing. The figures do not establish that every customer will pay $4.52 per task.
As an affiliate, we earn on qualifying purchases.
Two Models, Different Access
The release follows Muse Spark 1.3 by six weeks, according to the source report. The earlier model scored 48 on the Intelligence Index. The new API model scored 52, while Muse Wonder Full reached 56. This separation is central to understanding the release: the stronger score belongs to the model available only within Muse Code, not the model customers can purchase through Meta’s API.
Artificial Analysis also evaluated model-and-product combinations in its Coding Agents Index. Muse Code with Muse Wonder Full scored 68, tying Claude Code with Claude Sonnet 5.5 at 68. Claude Code with Claude Opus 5.5 scored 66, while Muse Code with Muse Wonder scored 63 and with Muse Spark 1.3 scored 54. Because this index assesses a model together with its software harness, the result compares those combinations; it does not establish how Wonder Full would perform inside Claude Code.
Other reported results show improvement alongside areas where the models remain behind some competitors. Muse Wonder Full scored 57% on Terminal-Bench 4.0, up from 33% for Spark 1.3, and 71% on AutomationBench-AA. The report gives it an AA-Omniscience score of 30, while Muse Wonder scored 25. Artificial Analysis reported hallucination rates of 23% for Full and 27% for Wonder, compared with 33% for Spark 1.3. Those figures are results on the evaluator’s tests and should not be read as universal rates across all uses.
“The capability gap closed. The price gap did not.”
— Thorsten Meyer, ThorstenMeyerAI.com
As an affiliate, we earn on qualifying purchases.
Costs Beyond the Benchmark
The source material does not provide a per-token price for Muse Wonder Full, so its task-level cost cannot be directly compared with the API model or rival services. It also does not identify a public, independently verifiable breakdown of Meta’s own billing for typical customer workloads. Artificial Analysis’ $4.52 figure is its estimate for a task in its benchmark, not a universal charge.
The supplied report gives only part of its cost breakdown, stating that input accounted for $3.43, or 76%, of Muse Wonder’s $4.52 task estimate. The remaining details are not included in the source material. The reasons for the models’ differing availability and the timing or scope of any broader release are also not specified.
Benchmark results cannot settle how the models will perform across different industries, prompt styles, or live coding projects. The report gives test scores and hallucination rates, but those measures do not establish reliability in every deployment. Users handling consequential work may still need review and verification processes.
As an affiliate, we earn on qualifying purchases.
Availability and Pricing Questions
The immediate next step for prospective users is to compare access, performance, and task-specific costs. Developers can access Muse Wonder through Meta’s API, while Muse Wonder Full is confined to Muse Code under the availability described in the report. Anyone assessing the API should test representative workloads and measure token consumption rather than treating the evaluator’s benchmark bill as a forecast for all use.
Further information from Meta about Muse Wonder Full’s pricing and availability would make a direct cost comparison possible. Until Meta publishes those details, the evaluation can establish the model’s benchmark results but not what it would cost to use at scale or whether Full will be offered through the API. No further release date or pricing announcement is provided in the source material.
As an affiliate, we earn on qualifying purchases.
Key Questions
What are Muse Wonder and Muse Wonder Full?
They are two proprietary Meta AI models. Both support text, image, audio and video input and have a one-million-token context window, according to the source report.
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 ten index evaluations; the result is a benchmark measurement, not a guarantee for every task.
Can customers buy Muse Wonder Full through Meta’s API?
Not according to the source material. It says Muse Wonder Full is available only through Muse Code; Muse Wonder is the model offered through Meta’s API.
How much does a Muse Wonder benchmark task cost?
Artificial Analysis estimates $4.52 per Intelligence Index task for Muse Wonder, based on the test’s token use and the model’s list prices. That is an evaluator calculation, not a guaranteed bill for a typical customer request.
How does Muse Wonder compare with GPT-6.1 Sol on cost?
In Artificial Analysis’ task-cost comparison, Muse Wonder scored 52 at an estimated $4.52 per task, while GPT-6.1 Sol also scored 52 at an estimated $0.72. The comparison applies to that benchmark and its estimated token use.
Source: ThorstenMeyerAI.com
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
