Muse Wonder Full vs Gemini 4 Carbon

Verdict

Gemini 4 Carbon scores higher — 60 at high vs 50 at max. One of the two has no cost figure from AA, so no price comparison is made.

Gemini 4 Carbon ranks #61 of 73 on cost per task among closed-source models above $1 per million tokens (blended 3:1) that AA prices; the class holds 237 models.

Our pick: Gemini 4 Carbon. The price sentences above use it as the baseline; scope: API access.

• Muse Wonder Full — Muse Wonder Full cannot be costed from AA: its AA page carries no per-token price and no per-task cost. This page therefore makes no claim about which row is cheaper. The axes both rows do carry: 50 vs 60 index points, the same four input modalities, and a 1M context window on both.
• Gemini 4 Carbon — Take Gemini 4 Carbon for API work as of this date: it is the only row of the two that AA prices — $7.76 per task, 25.7 minutes — and it scores 60 vs 50.

Side by side

Where each wins

Muse Wonder Full

• Not comparable on Cost per task (USD) (N/A — AA reports Unknown vs $7.76)
• Not comparable on Minutes per task (N/A — AA reports Unknown vs 25.7 min)
• Not comparable on Cost per index point (not computable (AA prints N/A for per-task cost) vs $0.1293)
• Not comparable on Input $/M tokens (not applicable — AA prints a dash vs $4.00)
• Not comparable on Output $/M tokens (not applicable — AA prints a dash vs $20.00)
• Not comparable on Cache discount (not applicable — no per-token price vs 95%)
• Level on Input modalities (text, image, speech, video)
• Level on Context window (1M)
• Lower on Intelligence Index

Gemini 4 Carbon

• Intelligence Index: 60 vs 50

Sources and conflicts

• Artificial Analysis lists Muse Wonder Full; OpenRouter's catalogue does not show it. This page reports what AA publishes and does not claim the model is absent.
• OpenRouter lists google/gemini-4-carbon (exact match). Input price: AA $4 per million tokens vs OpenRouter $2 — a 2.0× gap. This page uses AA's figure throughout and does not average the two.
• Prices and input modalities come from Artificial Analysis only; the second source is used to cross-check, never to union or average.
• A differing modality string is not a conflict: AA records what the model accepts, a provider records what its own route declares. Different objects, both reported here.

Data

Data as of 2026-11-12 · Artificial Analysis Intelligence Index v5.0.1, highest reasoning effort per model. Prices from the same pages.
Every cell is tied to a verbatim excerpt of that model's own Artificial Analysis page.
N/A (AA: Unknown) = AA publishes no value in that cell. It is not 0, and it is not 'unreleased'.
'not applicable (AA prints `-`)' = AA lists no per-token price for this model, so the cell does not exist. It is not the same as N/A and not the same as a withheld figure.
'not computable (…)' = the cell is derived from two others and one of them has no figure. The bracket names which one — it is not a value, and it is not the same as N/A.
'not printed on its AA page' = the page lists prices but withholds that figure; it is not treated as equal to anything.
'unverified' = we have not read that cell on that page. It is not a value and not a tie.
Token rows are AA's Verbosity block: input and output totals for the whole Intelligence Index run. 'Output/input ratio' is AA's own label and means output / input — it is not output's share of the total.
AA publishes no measured cache-hit rate for either model (it prints a fixed 7:2:1 cache-hit/input/output blend instead), so no cell records one.