by MetaCoding AgentfreemiumRank #3 on Leaderboard
Muse Code
Coding agent harness from Meta, measured on the Artificial Analysis Coding Agent Index.
Overall Index
54.5/ 100
Artificial Analysis benchmark compositeSWE-bench Verified
—
Real-world GitHub issue resolutionFrontierCode-AA 2.0
35.0%
Multi-file software engineering tasksTerminal-Bench 4.1
59.6%
Shell execution & command accuracyAvg Cost / Task
—
Inference token expenditure per taskOverview & Analysis
Muse Code was evaluated by Artificial Analysis with Muse Wonder Full (Max). We have not written a full profile for it yet.
Pricing Model: See the vendor for current pricing.
Technical Specs & Models
Muse Wonder Full (Max)
Key Capabilities & Features
Specialized engineering mechanisms implemented in Muse Code.
Pros & Cons Evaluation
Forensic analysis of developer experience, cost efficiency, and trade-offs.
Strengths & Advantages
Limitations & Trade-Offs
Pricing Plans & Tiers
Subscription plans, credit quotas, and commercial deployment options.
Verified Benchmark Evaluations
Standardized evaluation scores recorded on SWE-bench and synthetic code harnesses.
| Benchmark Harness | Score / Accuracy | Evaluation Metric | Verification Status | Evaluation Notes |
|---|---|---|---|---|
| Coding Agent Index (v2.1) | 54.5% | Composite of three benchmarks | Verified | Source: Artificial Analysis. |
| FrontierCode-AA 2.0 | 35% | Software engineering tasks | Verified | Standard benchmark instance run. |
| Terminal-Bench 4.1 | 59.6% | Agentic terminal use | Verified | Standard benchmark instance run. |
| SWE-Atlas-QnA-AA | 68.8% | Technical Q&A | Verified | Standard benchmark instance run. |