Skip to contentT/TTokens or Towers

Compare

LeaderboardEvery tracked model, ranked and priced side by sideBenchmarksWhat each test measures and who leads itPricesAPI, media, and hardware rates with a cost calculator

Plan

Stack plannerSix questions to a priced local and hosted stackShortlistOur picks for subscriptions, APIs, and hardwareEnterpriseSeat plans against self-hosting for a whole team

Build

HardwareGPUs and workstations, and which models they fitHostsWhere to rent open-weight inference, and for how muchHarnessesCoding agents and the models they drive

Lab

FindingsResults from our own evaluationsScanAudit a repo’s AI usage in the browserMethodWhere the numbers come from and how we check them
Data checked September 25, 2026
Data checked September 25, 2026
Plan a stack
T/TTokens or Towers

An independent guide to AI models, what they cost, and the hardware to run them. No account, no API key, no sponsors.

Compare

  • Leaderboard
  • Benchmarks
  • Prices

Plan

  • Stack planner
  • Shortlist
  • Enterprise

Build

  • Hardware
  • Hosts
  • Harnesses

Lab

  • Findings
  • Scan
  • Method
Data checked September 25, 2026 · Every number links to its sourceRSS · Source on GitHub
Leaderboard/Meta
Meta

Meta · Muse Wonder

Muse Wonder Full

Closed weightsPreviewNewproprietary

The larger Muse Wonder, available only to Muse Code subscribers ($50 Power and $200 Max plans) and not on the Meta Model API, so it has no token price.

AA Intelligence
56#2 of 151
Price
—
Output speed
—
Context window
1.05M tokens

Composite indexes

  • Artificial Analysis Intelligence Index#2 of 151
    56independentAA Intelligence Index v4.3.2, max effort; AA labels this an estimate pending independent evaluationArtificial Analysis: Muse Wonder Full (max) ↗

    Leader: Claude Opus 5.5 at 58

Coding

  • Terminal-Bench 4.0#6 of 96
    57%independentTerminal-Bench 4.0 (AA harness), max effortArtificial Analysis: Muse Wonder analysis ↗

    Leader: Claude Sonnet 5.5 at 70.6%

  • SciCode#3 of 100
    63%independentmax effortArtificial Analysis: Muse Wonder analysis ↗

    Leader: Claude Opus 5.5 at 66.9%

Agents and tool use

  • GDPval-AA#1 of 109
    1857independentGDPval-AA, max effortArtificial Analysis: Muse Wonder analysis ↗

    Leader: Muse Wonder Full at 1857

  • OSWorld-Verified#8 of 21
    78.4%vendorOSWorld V2.1, partial credit, max effort (binary 39.4)Meta: Introducing Muse Wonder ↗

    Leader: Kimi K3 at 84.8%

Reasoning and knowledge

  • CritPt#1 of 144
    33%independentmax effortArtificial Analysis: Muse Wonder analysis ↗

    Leader: Muse Wonder Full at 33%

Other published results

ResultScoreReported bySource
AA-Omniscience Index · max effort30 index (-100 to 100)independentArtificial Analysis: Muse Wonder analysis ↗
DeepSWE v1.1 · max effort79.0%vendorMeta: Introducing Muse Wonder ↗
SWE-Marathon · xhigh57.5%vendorMeta: Introducing Muse Wonder ↗
AutomationBench-AA · max effort45.4%vendorMeta: Introducing Muse Wonder ↗
Vals Finance Agent V2 · max effort59.1%vendorMeta: Introducing Muse Wonder ↗

Specs

Released
October 12, 2026
Parameters
Undisclosed
Size class
Undisclosed
Input
text, image, video, audio
Output
text
Context
1.05M tokens
Max output
Not published

    Will it fit your budget?

    The planner prices this model against local hardware and other APIs for your workload.

    Plan a stack →

    Compare with

    • AnthropicClaude Sonnet 5.556
    • AnthropicClaude Opus 5.557.6
    • AnthropicClaude Fable 5.153.4
    • GoogleGemini 4 Argon53