BenchLeader

GPT-6.2 Sol vs Muse Spark

Full profiles: GPT-6.2 Sol (OpenAI) and Muse Spark (Meta): every benchmark, price and rank.

Updated · Data from 10 sources

As of 19 Nov 2026, GPT-6.2 Sol and Muse Spark are neck and neck on quality (64.6 vs 64.1 on the BenchLeader Index, inside the ±4.6 margin). GPT-6.2 Sol leads in every category we measure both in.

Key facts (as of 19 Nov 2026)

  • GPT-6.2 Sol scores 64.6 on the BenchLeader Index and Muse Spark scores 64.1 (BenchLeader, 19 Nov 2026).
  • GPT-6.2 Sol costs $4.00 per 1M tokens blended (list price, 19 Nov 2026).
  • Of the 2 benchmarks both models have, GPT-6.2 Sol leads 2 (BenchLeader, 19 Nov 2026).

Which to pick for…

The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.

Use casePickWhyConfidence
CodingGPT-6.2 Sol75 vs 65 category score; Vibe Code Bench v1.1 87.6% vs 19.7%High
Agents & tool useEither65 vs 65 category score; AA-Briefcase v1.1 1612 vs 646Either
ReasoningGPT-6.2 Sol70 vs 67 category scoreMedium
KnowledgeGPT-6.2 Sol65 vs 63 category scoreMedium
Chat quality (human preference)–Only Muse Spark has an Arena rating (68 category score)Not comparable
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • GPT-6.2 Sol (max) and Muse Spark are neck and neck on quality (64.6 vs 64.1, within the ±4.6 margin).
  • GPT-6.2 Sol (max) is stronger in agents & tools, coding, knowledge, reasoning.

At a glance

SpecGPT-6.2 SolMuse Spark
DeveloperOpenAIMeta
Released17 Nov 20268 Apr 2026
Context window (tokens)1,050,000262,144
Compared atGPT-6.2 Sol (max)Muse Spark
Open weightsNoNo
Price in / out / cached input per 1M$2.00 / $10.00 / $0.100–

Category by category

Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.

CategoryGPT-6.2 Sol (max)Muse SparkLeader
Reasoning70.167.4GPT-6.2 Sol
Coding74.565.1GPT-6.2 Sol
Agents & tools65.165.0level
Maths–59.1not comparable
Knowledge65.462.8GPT-6.2 Sol
Instruction following–76.2not comparable
Human preference–68.1not comparable
Multimodal–64.0not comparable
Long context–63.6not comparable
Composite64.5–not comparable

On one chart

Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.

2550 avg75ReasoningCodingAgents & toolsKnowledgeGPT-6.2 Sol, Reasoning: 70.1GPT-6.2 Sol, Coding: 74.5GPT-6.2 Sol, Agents & tools: 65.1GPT-6.2 Sol, Knowledge: 65.4Muse Spark, Reasoning: 67.4Muse Spark, Coding: 65.1Muse Spark, Agents & tools: 65.0Muse Spark, Knowledge: 62.8

Not drawn, measured for fewer than two of these models: Maths, Instruction following, Long context, Multimodal, Human preference.

Model, click to add or removeIndex

Where sources disagree

Across the 2 benchmarks both have, GPT-6.2 Sol leads 2, Muse Spark leads 0.

Overall indexGPT-6.2 Sol (max)Muse SparkLeader
BenchLeader Index64.664.1neck and neck

Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).

Price, speed and latency

List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.

MeasureGPT-6.2 Sol (max)Muse Spark
Blended price per 1M tokens$4.00–
Input / output per 1M$2.00 / $10.00– / –
Cached input per 1M$0.100–
Price per task: chat reply$0.0038–
Price per task: summarize a 30-page report$0.030–
Price per task: code edit$0.027–
Price per task: agentic coding session$0.160–
Price per task: structured extraction$0.0060–
Agentic coding session, input mostly cached$0.074–
Cost per benchmark task (AA Index)$1.01–
Output tokens per task (AA Index)40k–
Of which reasoning23k–
Time per task (AA Index)9 min–
Output speed74 tok/s–
Time to first answer78.8 s–

Each model is compared at its best-ranked reasoning-effort setting. GPT-6.2 Sol (max) scores 64.6 and takes 79 s to its first answer token; GPT-6.2 Sol (xhigh) scores 63.3 in 32 s, GPT-6.2 Sol (high) scores 62.3 in 12 s. Muse Spark scores 64.1. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.

Every benchmark

113 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.

MetricPublisherGPT-6.2 Sol (max)Muse Spark
BenchLeader IndexBenchLeader64.664.1
Agents & tools scoreBenchLeader65.165.0
Coding scoreBenchLeader74.565.1
Composite scoreBenchLeader64.5–
Knowledge scoreBenchLeader65.462.8
Reasoning scoreBenchLeader70.167.4
Human preference scoreBenchLeader–68.1
Instruction following scoreBenchLeader–76.2
Long context scoreBenchLeader–63.6
Maths scoreBenchLeader–59.1
Multimodal scoreBenchLeader–64.0
Blended price $/MBenchLeader$4.00–
Output speedBenchLeader74 tok/s–
Time to first answerBenchLeader78.8 s–
Context windowBenchLeader1,050,000 tokens262,144 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub–89.8%
Humanity's Last ExamScale AI / CAIS–40.6%
Arena Hard PromptsArena (LMArena)–1506
LiveBench 2.0 ReasoningLiveBench90.6%–
GPQA Diamond (AA)Artificial Analysis–88.4%
Humanity's Last Exam (AA)Artificial Analysis–40.7%
HLE-Diamond (AA)Artificial Analysis64.0%–
GPQA Diamond (Vals)Vals AI–89.7%
Kagi LLM BenchmarkKagi LLM Benchmark67.0%†–
ARC-AGI-1ARC Prize97.5%–
ARC-AGI-2ARC Prize85.8%–
ARC-AGI-3ARC Prize98.2%–
CritPtArtificial Analysis–11.3%
CritPt v2 hard subset (AA)Artificial Analysis62.5%–
MysteryMechanismVals AI36.9%–
HLE-Diamond (Scale)Scale AI SEAL56.5%†–
Arena ExpertArena (LMArena)–1490
Coding
SciCodeSciCode–51.5%
Arena CodingArena (LMArena)–1529
Arena WebDevArena (LMArena)1757–
LiveBench 2.0 CodingLiveBench83.7%–
SWE-bench (Vals)Vals AI–74.4%
SWE-Bench ProScale AI SEAL–55.0%
Vibe Code Bench v1.1Vals AI87.6%19.7%
Vibe Code Bench 1-100 v1.1Vals AI40.9%–
Show all 113 benchmarks
MetricPublisherGPT-6.2 Sol (max)Muse Spark
Coding
SWE-Bench Pro (private)Scale AI SEAL–44.7%
SWE Atlas v1.0: Codebase QnAScale AI SEAL–24.2%
SWE Atlas v1.0: Test WritingScale AI SEAL–31.1%
SWE Atlas v1.1Scale AI SEAL60.5%†–
SWE Atlas v1.1: Codebase QnAScale AI SEAL61.3%†–
SWE Atlas v1.1: Test WritingScale AI SEAL64.4%†–
SWE Atlas v1.1: RefactoringScale AI SEAL55.7%†–
Terminal-Bench 4.0 (AA)Artificial Analysis60.1%–
Terminal-Bench 2.1 (AA)Artificial Analysis–62.2%
FrontierCode 2.0 (AA)Artificial Analysis44.4%–
Agents & tools
Terminal-BenchTerminal-Bench61.5%–
Arena AgentArena (LMArena)11.3–
LiveBench Long AgenticLiveBench63.6%–
LiveBench BrowserLiveBench88.2%–
Terminal-Bench HardArtificial Analysis–45.5%
τ²-Bench Telecom (AA)Artificial Analysis–91.5%
MCP AtlasScale AI SEAL–82.2%
GDPval-AA v2.1Artificial Analysis–1006
Analyst Agent (AA)Artificial Analysis53.8%–
Finance Agent v2.1Vals AI58.9%–
Harvey's Legal Agent Benchmark v1.1Vals AI12.1%–
Legal Research BenchVals AI37.0%–
Tax Agent BenchVals AI24.8%–
Terminal-Bench 2.0 (Vals)Vals AI–59.5%
Terminal-Bench 4.0 (Vals)Vals AI60.6%–
Terminal-Bench ScienceVals AI58.6%–
Arena Agent Task CompletionArena (LMArena)10.9–
AutomationBenchArtificial Analysis71.3%–
GDP.pdf (AA)Artificial Analysis30.0%–
Harvey LAB-AA v1.1Harvey11.1%–
EnterpriseOps-GymArtificial Analysis28.9%–
AA-Briefcase v1.1Artificial Analysis1612646
Terminal-Bench-Science 0.1 (AA)Artificial Analysis67.6%–
CWE-Bench (AA)Artificial Analysis61.7%–
DeepSec-Bench (AA)Artificial Analysis51.6%–
CyberGym end-to-end (AA)Artificial Analysis0.0%–
Maths
OTIS Mock AIMEEpoch AI Benchmarking Hub–88.9%
ProofBenchVals AI–17.0%
AIME (Vals)Vals AI–96.9%
Autoform-HardVals AI85.7%–
Arena MathsArena (LMArena)–1468
Knowledge
LiveBench 2.0 Data AnalysisLiveBench81.1%–
LiveBench ResearchLiveBench90.1%–
AA-Omniscience v1.1Artificial Analysis40.4–
MMLU-ProVals AI–87.3%
LegalBenchVals AI–84.2%
CorpFinVals AI–65.1%
TaxEvalVals AI–77.7%
PRBench FinanceScale AI SEAL–52.4%
PRBench LegalScale AI SEAL43.3%†52.3%
MultiNRCScale AI SEAL–59.0%
BioMysteryBenchVals AI77.8%–
CaseLaw v2Vals AI–63.1%
Excel Modeling Benchmark v1.1Vals AI74.3%–
MedCodeVals AI–51.3%
MedScribeVals AI–85.9%
DrugDiscoveryBenchScale AI SEAL62.2%†–
AA-Omniscience v1.1: accuracyArtificial Analysis61.6%–
AA-Omniscience v1.1: non-hallucinationArtificial Analysis44.6%–
Instruction following
IFBenchArtificial Analysis–75.9%
MultiChallengeScale AI SEAL72.0%†75.5%
TutorBenchScale AI SEAL–68.5%
Arena Instruction FollowingArena (LMArena)–1465
Human preference
Arena TextArena (LMArena)–1489
Arena Creative WritingArena (LMArena)–1465
Arena Multi-turnArena (LMArena)–1493
Arena Longer QueriesArena (LMArena)–1476
Multimodal
Arena VisionArena (LMArena)–1306
LiveBench DesignLiveBench80.0%–
MMMU-ProArtificial Analysis–80.5%
MMMU-Pro (Vals)Vals AI–87.4%
Humanity's Sixth SenseScale AI SEAL56.8%†–
VIALS-AAHandshake AI62.7%–
AA-3DArtificial Analysis1259–
Long context
AA-LCR v1.1Artificial Analysis–78.0%
Arena DocumentArena (LMArena)–1467
MLCRArtificial Analysis37.2%–
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub–153.2
LiveBench 2.0LiveBench80.3%–
AA Intelligence Index v5.0.1Artificial Analysis57.3–
AA Intelligence Index v4.3.2Artificial Analysis–31.3
AA Progress Index v1Artificial Analysis–289.1
Vals IndexVals AI66.1–
Vals Finance IndexVals AI56.7%–
Vals Science IndexVals AI57.8%–
Vals Legal IndexVals AI66.7%–
AA Cyber IndexArtificial Analysis37.8–
Safety & honesty
FORTRESSScale AI SEAL–20.2%

Data as of 2026-11-19. Best configuration of each model; every score links to its source on the model pages.

What changed

  • GPT-6.2 Sol (max): BenchLeader Index 63.2 → 64.6.
  • GPT-6.2 Sol (max): 4 new results (VIALS-AA, Arena WebDev, Arena Agent…).
  • GPT-6.2 Sol (max): BenchLeader Index 62.4 → 63.2.
  • GPT-6.2 Sol (max): 2 new results (Vibe Code Bench 1-100 v1.1, Terminal-Bench).
  • Muse Spark: 2 new results (GDPval-AA v2.1, AA-Briefcase v1.1).
  • Muse Spark: BenchLeader Index 64.6 → 64.1.
  • Muse Spark: 1 score revision.
  • Muse Spark: 1 new result (AA Progress Index v1).

Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs

GPT-6.2 Sol vs Muse Spark: questions

Is GPT-6.2 Sol better than Muse Spark?
GPT-6.2 Sol (max) and Muse Spark are neck and neck on quality (64.6 vs 64.1, within the ±4.6 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 19 Nov 2026, so check the category scores for your use.
Is GPT-6.2 Sol or Muse Spark better for coding?
GPT-6.2 Sol scores higher in coding (75 vs 65 on the category index, where 50 is average).
Is GPT-6.2 Sol or Muse Spark better for agents and tool use?
They are level in agentic tasks (65 vs 65 on the category index).
Which is better for long documents?
GPT-6.2 Sol accepts more context: 1,050,000 against 262,144 tokens for Muse Spark.
Do independent benchmarks agree on GPT-6.2 Sol vs Muse Spark?
Across the 2 benchmarks both have, GPT-6.2 Sol leads 2, Muse Spark leads 0.
More questions (1)
When was GPT-6.2 Sol released?
GPT-6.2 Sol was released on 17 Nov 2026.