Muse Wonder

Where people rank it

From blind votes on LMArena, where people pick the better of two anonymous answers. See the full tables on the AI leaderboard.

On paper

Fixed-test rankings from the benchmarks tab: Speed #2, Reasoning #7, Value #12.

Strengths and weaknesses

Best for

• Agent and coding work in Muse Code
• Fast frontier-level answers
• General chat

Every score we have

Each bar compares this model with the best result any model on this page has on the same test.

• LMArena text arena (human votes)1,508.9Arena score 1508.9 (95% CI 1498.3 to 1519.5); 3,247 votes; entry "Muse Wonder (Max)"; position 2 of 416 on the full board
• LMArena creative writing1,485.1Arena score 1485.1 (95% CI 1461.9 to 1508.2); 710 votes; entry "Muse Wonder (Max)"; position 8 of 414 on the full board
• LMArena WebDev (coding)1,650.4Arena score 1650.4 (95% CI 1635.7 to 1665.1); 1,942 votes; entry "Muse Wonder (Max)"; position 15 of 144 on the full board
• LMArena Agent Arena0.049IPS score 0.0494 (95% CI 0.0323 to 0.0664); 7,690 sessions; entry "Muse Wonder (Max)"; position 11 of 52 on the full board
• Artificial Analysis Intelligence Index52.452.4 on the Artificial Analysis Intelligence Index v4.3.2 (AA shows 52), setting "Muse Wonder (max)"
• Humanity's Last Exam (Artificial Analysis)50.1%50.1% on Humanity's Last Exam (AA, 2,158 text-only questions), setting "Muse Wonder (max)"
• Terminal-Bench 4.0 (Artificial Analysis run)52.0%52.0% on Terminal-Bench 4.0 as run by AA (mini-swe-agent harness), setting "Muse Wonder (max)"
• Long-context reasoning (AA-LCR)80.7%80.7% on AA-LCR (long-context reasoning), setting "Muse Wonder (max)"
• AA-Omniscience, accuracy minus hallucination25.525.5 on the AA-Omniscience Index (knowledge accuracy minus hallucination, range -100 to 100), setting "Muse Wonder (max)"
• Output speed, tokens per second181 t/s181 output tokens per second (AA median), row "Muse Wonder (max)"
• Cost per Artificial Analysis test task$4.52$4.52 to run one AA Intelligence Index task, row "Muse Wonder (max)"
• Artificial Analysis Coding Agent Index63.563.5 on the AA Coding Agent Index v1.5 (DeepSWE v1.1, Terminal-Bench 4.0, SWE-Atlas-QnA), best listed setup "Muse Code - Muse Wonder (max)"