Enter opens the highlighted result. Esc closes.

GPT-6.2 Sol vs Muse Wonder

GPT-6.2 Sol leads Muse Wonder on coding, 83.1 to 71.8 on the Vibe Coding Index, and GPT-6.2 Sol costs 2.0x less per output token ($10.00 vs $20.00 per million). Updated November 22, 2026; 41 benchmarks have a result for both.

GPT-6.2 SolOpenAI
Coding
83.1
Overall
79.9
Output $/M
$10.00
21
Muse WonderMeta
Coding
71.8
Overall
72.6
Output $/M
$20.00
20
vs boards won of 41 shared

GPT-6.2 Sol or Muse Wonder: which to pick

Pick GPT-6.2 Sol for coding: it scores 83.1 on the Vibe Coding Index against 71.8 for Muse Wonder.

Pick Muse Wonder for work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.

Shared results
41
GPT-6.2 Sol only
15
Muse Wonder only
41
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
GPT-6.2 Sol 83.1 from 4 benchmarks, Muse Wonder 71.8 from 3 benchmarks.
Widest category
Math, 39.6 points apart on average over 1 shared benchmark.
Widest single gap
MathArena: GPT-6.2 Sol 91.2% against Muse Wonder 51.6%, 39.6 points apart on a 0 to 100 scale.
Output price
GPT-6.2 Sol $10.00 against Muse Wonder $20.00 per million output tokens, 2.0x apart.
Context window
GPT-6.2 Sol takes 1.05M tokens, Muse Wonder 1.05M.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryGPT-6.2 SolBarsMuse WonderTies
Coding82
Intelligence10
Agentic work33
Reasoning32
Math10
Knowledge12
Web design30
General111
All shared boards2120
CodingLike-for-likeSix of these boards feed the Vibe Coding Index82
Coding benchmarks for GPT-6.2 Sol and Muse Wonder
BenchmarkGPT-6.2 SolMuse WonderLeader
LiveBench coding83.7%Not measuredOnly GPT-6.2 Sol measured
LMArena WebDev17541647GPT-6.2 Sol
Terminal-Bench 4 (Vals AI)60.6%49.5%GPT-6.2 Sol
Vibe Code Bench (Vals AI)87.6%83.1%GPT-6.2 Sol
Code migration (Vals AI)Not measured62.5%Only Muse Wonder measured
IOI (Vals AI)Not measured66.9%Only Muse Wonder measured
SWE-bench Multimodal (Vals AI)Not measured45.6%Only Muse Wonder measured
Vibe Code Bench 1-100 (Vals AI)Not measured26.6%Only Muse Wonder measured
CyberBench (Vals AI)Not measured80.4%Only Muse Wonder measured
SciCodeNot measured60.3%Only Muse Wonder measured
Terminal-Bench 4.0 (Artificial Analysis run)60.1%52.0%GPT-6.2 Sol
FrontierCode 2.044.4%27.9%GPT-6.2 Sol
LMArena Text: coding14401523Muse Wonder
LMArena Text: software and IT14341526Muse Wonder
LMArena WebDev: React17611655GPT-6.2 Sol
LMArena WebDev: HTML18401652GPT-6.2 Sol
GameBench 294.477.7GPT-6.2 Sol
Terminal-Bench 4.0Not measured47.0%Only Muse Wonder measured
ProgramBenchNot measured5.5%Only Muse Wonder measured
Neither model has a result on the other 33 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for GPT-6.2 Sol and Muse Wonder
BenchmarkGPT-6.2 SolMuse WonderLeader
AA Intelligence Index57.343.6GPT-6.2 Sol
Epoch Capabilities IndexNot measured161.7Only Muse Wonder measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index33
Agentic work benchmarks for GPT-6.2 Sol and Muse Wonder
BenchmarkGPT-6.2 SolMuse WonderLeader
Design Arena: agents building mobile appsNot measured1239Only Muse Wonder measured
Design Arena: HTML slidesNot measured1129Only Muse Wonder measured
Design Arena: PowerPoint slidesNot measured1034Only Muse Wonder measured
Design Arena: spreadsheetsNot measured1230Only Muse Wonder measured
LiveBench research90.1%Not measuredOnly GPT-6.2 Sol measured
LiveBench browser88.2%Not measuredOnly GPT-6.2 Sol measured
LiveBench long agentic63.6%Not measuredOnly GPT-6.2 Sol measured
Finance Agent (Vals AI)Not measured56.6%Only Muse Wonder measured
Tax Agent Bench (Vals AI)24.8%36.3%Muse Wonder
Legal Research Bench (Vals AI)37.0%50.5%Muse Wonder
Harvey's Legal Agent Benchmark (Vals AI)12.1%23.8%Muse Wonder
Excel Modeling Benchmark (Vals AI)Not measured67.1%Only Muse Wonder measured
BioMysteryBench (Vals AI)77.8%67.0%GPT-6.2 Sol
Terminal-Bench-Science (Vals AI)58.6%22.9%GPT-6.2 Sol
Vals Science Index57.8%39.1%GPT-6.2 Sol
CUA-bench (Vals AI)Not measured7.2%Only Muse Wonder measured
APEX-Agents66.6%Not measuredOnly GPT-6.2 Sol measured
PostTrainBenchNot measured24.2%Only Muse Wonder measured
Vending-Bench 2Not measured$9,432Only Muse Wonder measured
LMArena Search1179Not measuredOnly GPT-6.2 Sol measured
Neither model has a result on the other 24 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index32
Reasoning benchmarks for GPT-6.2 Sol and Muse Wonder
BenchmarkGPT-6.2 SolMuse WonderLeader
LiveBench reasoning90.6%Not measuredOnly GPT-6.2 Sol measured
MysteryMechanism (Vals AI)36.9%27.5%GPT-6.2 Sol
Humanity's Last Exam (Artificial Analysis run)Not measured50.1%Only Muse Wonder measured
HLE-Diamond (Artificial Analysis run)64.0%39.0%GPT-6.2 Sol
CritPt v2, hard subset (Artificial Analysis run)64.0%50.5%GPT-6.2 Sol
ARC-AGI97.5%Not measuredOnly GPT-6.2 Sol measured
ARC-AGI-285.8%Not measuredOnly GPT-6.2 Sol measured
SimpleBench69.9%Not measuredOnly GPT-6.2 Sol measured
LMArena Text: hard prompts14191514Muse Wonder
LMArena Text: expertNot measured1516Only Muse Wonder measured
LMArena Text: hard prompts in English14131519Muse Wonder
Kagi LLM Benchmark67.0%Not measuredOnly GPT-6.2 Sol measured
Humanity's Last Exam (Scale run)56.5%Not measuredOnly GPT-6.2 Sol measured
Neither model has a result on the other 13 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index10
Math benchmarks for GPT-6.2 Sol and Muse Wonder
BenchmarkGPT-6.2 SolMuse WonderLeader
ProofBenchNot measured84.0%Only Muse Wonder measured
LMArena Text: mathNot measured1493Only Muse Wonder measured
LMArena Text: mathematical workNot measured1509Only Muse Wonder measured
MathArena91.2%51.6%GPT-6.2 Sol
Neither model has a result on the other 10 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index12
Knowledge benchmarks for GPT-6.2 Sol and Muse Wonder
BenchmarkGPT-6.2 SolMuse WonderLeader
AA-Omniscience40.425.1GPT-6.2 Sol
LMArena Text: science14321507Muse Wonder
LMArena Text: medicine and healthcareNot measured1500Only Muse Wonder measured
LMArena Text: legal and governmentNot measured1514Only Muse Wonder measured
LMArena Text: business and finance13831490Muse Wonder
Neither model has a result on the other 16 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index30
Web design benchmarks for GPT-6.2 Sol and Muse Wonder
BenchmarkGPT-6.2 SolMuse WonderLeader
Design Arena: websitesNot measured1374Only Muse Wonder measured
Design Arena: UI componentsNot measured1357Only Muse Wonder measured
Design Arena: game dev14371394GPT-6.2 Sol
Design Arena: data vizNot measured1290Only Muse Wonder measured
Design Arena: 3D14821420GPT-6.2 Sol
Design Arena: all code categories14051384GPT-6.2 Sol
Design Arena: SVGNot measured1286Only Muse Wonder measured
LiveBench design80.0%Not measuredOnly GPT-6.2 Sol measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index111
General benchmarks for GPT-6.2 Sol and Muse Wonder
BenchmarkGPT-6.2 SolMuse WonderLeader
Vals Index66.1%64.9%GPT-6.2 Sol
LiveBench data analysis81.1%Not measuredOnly GPT-6.2 Sol measured
Public Benefits Bench (Vals AI)Not measured70.2%Only Muse Wonder measured
Long context reasoning (AA-LCR)Not measured80.7%Only Muse Wonder measured
GDPval-AANot measured1770Only Muse Wonder measured
DTBenchNot measured95.2%Only Muse Wonder measured
LMCANot measured56.8%Only Muse Wonder measured
LMArena Text14141503Muse Wonder
LMArena Text: instruction following14341492Muse Wonder
LMArena Text: creative writing14031488Muse Wonder
LMArena Text: English14191509Muse Wonder
LMArena Text: non-English14051492Muse Wonder
LMArena Text: Chinese14341542Muse Wonder
LMArena Text: GermanNot measured1508Only Muse Wonder measured
LMArena Text: JapaneseNot measured1499Only Muse Wonder measured
LMArena Text: KoreanNot measured1465Only Muse Wonder measured
LMArena Text: FrenchNot measured1515Only Muse Wonder measured
LMArena Text: SpanishNot measured1480Only Muse Wonder measured
LMArena Text: Russian14011511Muse Wonder
LMArena Text: PolishNot measured1535Only Muse Wonder measured
LMArena Text: multi-turn14201512Muse Wonder
LMArena Text: longer queries14271499Muse Wonder
LMArena Text: writing and language14201494Muse Wonder
LMArena Text: entertainment and media14081481Muse Wonder
MultiChallenge72.0%Not measuredOnly GPT-6.2 Sol measured
EQ-Bench 4Not measured1325Only Muse Wonder measured
Neither model has a result on the other 23 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-6.2 Sol and Muse Wonder
BenchmarkGPT-6.2 SolMuse WonderLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

GPT-6.2 Sol$0.0070Fits in one request
Muse Wonder$0.014Fits in one request

GPT-6.2 Sol costs less

Repository review

50K input + 3K output tokens

GPT-6.2 Sol$0.130Fits in one request
Muse Wonder$0.260Fits in one request

GPT-6.2 Sol costs less

Agent loop

200K input + 10K output tokens

GPT-6.2 Sol$0.500Fits in one request
Muse Wonder$1.00Fits in one request

GPT-6.2 Sol costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of GPT-6.2 Sol and Muse Wonder
SpecificationOpenAI GPT-6.2 SolMeta Muse Wonder
MakerOpenAIMeta
WeightsNo public weights listedNo public weights listed
ReleasedNovember 17, 2026October 12, 2026
Context window1.05M1.05M
Input $/Mtok$2.00$4.00
Output $/Mtok$10.00$20.00
Coding index83.171.8
Independent measurements behind it43
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-6.2 Sol better than Muse Wonder for coding?

Pick GPT-6.2 Sol for coding: it scores 83.1 on the Vibe Coding Index against 71.8 for Muse Wonder. Pick Muse Wonder for work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.

Which is cheaper, GPT-6.2 Sol or Muse Wonder?

GPT-6.2 Sol costs $2.00 per million input tokens and $10.00 per million output tokens; Muse Wonder costs $4.00 and $20.00. Cost per solved task is known for neither of them.

Which has the bigger context window, GPT-6.2 Sol or Muse Wonder?

GPT-6.2 Sol, with 1.05M tokens against 1.05M.

How many benchmarks is the GPT-6.2 Sol vs Muse Wonder comparison based on?

41 benchmarks have a result for both models as of November 22, 2026. GPT-6.2 Sol's coding index rests on 4 independent benchmarks and Muse Wonder's on 3. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/gpt-6-2-sol-vs-meta-muse-wonder" title="GPT-6.2 Sol vs Muse Wonder, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/gpt-6-2-sol-vs-meta-muse-wonder">LLMBenchmark.ai, GPT-6.2 Sol vs Muse Wonder</a></p>