Inference
EN

GPT-5.6 Terra vs Muse Wonder Full

Change models
GPT-5.6 TerraOpenAI
Muse Wonder FullMeta
IndexAll categories, out of 100.
68.9 ±2.0
77.3 ±5.3 best
GapIndex points below the top.
−8.4
Top
CoverageIndex weight that has results.
100% best
80%
SpeedOutput tokens per second.
66/s
—
EloLMArena rating and rank.
1449
—
ReleasedMonth of release.
2026-07
2026-10
Input / 1MUS dollars per 1M input tokens.
$2.00
—
Output / 1MUS dollars per 1M output tokens.
$12.00
—
Batch output / 1MBatch price per 1M output tokens.
$6.00
—
1K chats2K tokens in, 500 out, each.
$10.00
—
1K documents20K tokens in, 1K out, each.
$52.00
—
1K agent tasks200K tokens in, 20K out, each.
$640.00
—
ContextMaximum tokens in one request.
1.05M best
1M
AgenticMulti-step tasks with tools.
68.2
78.2 best
CodingCode writing and repair.
68.4
77.3 best
ReasoningLogic problems and puzzles.
72.5
75.6 best
MultimodalTasks with images and text.
67.3
—
KnowledgeFacts and expert knowledge.
65.5
71.7 best
MultilingualTasks in many languages.
60.7
—
InstructionTasks with strict prompt rules.
56.4
—
MathMath problems.
72.8
79.6 best

Benchmarks

29 shared
Only GPT-5.6 Terra: 80Only Muse Wonder Full: 13
GPT-5.6 Terra
Muse Wonder Full
Artificial Analysis Humanity's Last ExamKnowledge
42.9%34 / 222
53.6% best9 / 222
Artificial Analysis Long Context ReasoningReasoning
83.0%18 / 217
83.7% best10 / 217
Critical Physics TasksReasoning
30.0%10 / 214
32.9% best2 / 214
Artificial Analysis Omniscience AccuracyKnowledge
46.8% best31 / 164
41.0%46 / 164
LegalBenchLegal
85.1%25 / 149
87.4% best4 / 149
GDPval-AA normalizedAgentic
47.6%33 / 133
67.9% best2 / 133
Vibe Code Bench v1.1Coding
74.6%44 / 123
86.0% best14 / 123
Artificial Analysis SciCodeCoding
55.0%29 / 114
62.6% best4 / 114
MedScribeHealthcare
82.9%50 / 109
89.3% best8 / 109
MedCodeHealthcare
43.4%45 / 106
49.9% best18 / 106
Legal Research BenchLegal
41.3%30 / 90
53.4% best6 / 90
Tax Agent BenchFinance
30.0%34 / 82
43.0% best8 / 82
Finance Agent (v2)Finance
54.4%27 / 79
59.1% best6 / 79
Code MigrationCoding
47.8%13 / 77
61.0% best7 / 77
EMBFinance
66.2%23 / 75
71.9% best10 / 75
ProgramBenchCoding
0.5%19 / 65
0.5%19 / 65
Terminal-Bench 4.0Agentic
22.7%32 / 61
58.1% best6 / 61
DeepSWEAgentic
69.6%23 / 59
79.0% best1 / 59
Agent Arena steerabilityAgentic
−0.527 / 54
9.6 best12 / 54
Agent Arena task outcomeAgentic
−7.241 / 54
11.1 best9 / 54