BenchLeader

GPT-6.1 Sol vs GPT-6.2 Astra

Full profiles: GPT-6.1 Sol (OpenAI) and GPT-6.2 Astra (OpenAI): every benchmark, price and rank.

Updated · Data from 14 sources

As of 19 Nov 2026, GPT-6.2 Astra is ahead of GPT-6.1 Sol on quality (68.9 vs 65.9 on the BenchLeader Index), but GPT-6.1 Sol is 4.0× cheaper at $4.00 vs $16.00 per 1M tokens blended. GPT-6.2 Astra leads in every category we measure both in.

Key facts (as of 19 Nov 2026)

  • GPT-6.1 Sol scores 65.9 on the BenchLeader Index and GPT-6.2 Astra scores 68.9 (BenchLeader, 19 Nov 2026).
  • GPT-6.1 Sol costs $4.00 per 1M tokens blended (list price, 19 Nov 2026).
  • GPT-6.2 Astra costs $16.00 per 1M tokens blended (list price, 19 Nov 2026).
  • Of the 61 benchmarks both models have, GPT-6.2 Astra leads 52 (BenchLeader, 19 Nov 2026).
  • On Arena WebDev, GPT-6.1 Sol is rated 1744 and GPT-6.2 Astra 1781 (LMArena, as published by 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 Astra76 vs 70 category score; Arena WebDev 1781 vs 1744Medium
Agents & tool useGPT-6.2 Astra66 vs 63 category score; Terminal-Bench 62.1% vs 60.9%Medium
MathsGPT-6.2 Astra73 vs 71 category score; FrontierMath Tiers 1–3 92.6% vs 93.7%Medium
ReasoningGPT-6.2 Astra73 vs 70 category score; GPQA Diamond 96.8% vs 95.4%Medium
KnowledgeGPT-6.2 Astra74 vs 70 category score; SimpleQA Verified 78.5% vs 73.9%Medium
Images and other mediaGPT-6.2 Astra72 vs 66 category score; MMMU-Pro 89.5% vs 86.0%Medium
Chat quality (human preference)–Only GPT-6.1 Sol has an Arena rating (67 category score)Not comparable
Lowest costGPT-6.1 Sol$4.00 vs $16.00 per 1M blended; agentic coding session $0.160 vs $0.640–
Fastest first answerGPT-6.1 Sol134.5 s (GPT-6.1 Sol (max)) vs 202.7 s (GPT-6.2 Astra (max)) to the first answer token–
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • GPT-6.2 Astra (max) leads on quality: 68.9 vs 65.9.
  • GPT-6.2 Astra (max) is stronger in agents & tools, coding, composite, knowledge, maths, multimodal, reasoning.
  • GPT-6.1 Sol (max) is 4.0× cheaper ($4.00 vs $16.00 per 1M blended).

At a glance

SpecGPT-6.1 SolGPT-6.2 Astra
DeveloperOpenAIOpenAI
Released29 Sep 202617 Nov 2026
Context window (tokens)1,050,0001,050,000
Compared atGPT-6.1 Sol (max)GPT-6.2 Astra (max)
Open weightsNoNo
Price in / out / cached input per 1M$2.00 / $10.00 / $0.100$8.00 / $40.00 / $0.400

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-6.1 Sol (max)GPT-6.2 Astra (max)Leader
Reasoning70.173.1GPT-6.2 Astra
Coding70.176.1GPT-6.2 Astra
Agents & tools63.166.4GPT-6.2 Astra
Maths71.372.5GPT-6.2 Astra
Knowledge70.374.2GPT-6.2 Astra
Human preference67.3–not comparable
Multimodal65.571.7GPT-6.2 Astra
Long context66.2–not comparable
Composite65.569.5GPT-6.2 Astra

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 & toolsMathsKnowledgeMultimodalGPT-6.1 Sol, Reasoning: 70.1GPT-6.1 Sol, Coding: 70.1GPT-6.1 Sol, Agents & tools: 63.1GPT-6.1 Sol, Maths: 71.3GPT-6.1 Sol, Knowledge: 70.3GPT-6.1 Sol, Multimodal: 65.5GPT-6.2 Astra, Reasoning: 73.1GPT-6.2 Astra, Coding: 76.1GPT-6.2 Astra, Agents & tools: 66.4GPT-6.2 Astra, Maths: 72.5GPT-6.2 Astra, Knowledge: 74.2GPT-6.2 Astra, Multimodal: 71.7

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

Model, click to add or removeIndex

Where sources disagree

Across the 61 benchmarks both have, GPT-6.1 Sol leads 7, GPT-6.2 Astra leads 52 and 2 are tied. Every overall index that rates both puts GPT-6.2 Astra ahead.

Overall indexGPT-6.1 Sol (max)GPT-6.2 Astra (max)Leader
BenchLeader Index65.968.9GPT-6.2 Astra
AA Intelligence Index v5.0.154.662.1GPT-6.2 Astra
Vals Index63.967.7GPT-6.2 Astra

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-6.1 Sol (max)GPT-6.2 Astra (max)
Blended price per 1M tokens$4.00$16.00
Input / output per 1M$2.00 / $10.00$8.00 / $40.00
Cached input per 1M$0.100$0.400
Price per task: chat reply$0.0038$0.015
Price per task: summarize a 30-page report$0.030$0.120
Price per task: code edit$0.027$0.108
Price per task: agentic coding session$0.160$0.640
Price per task: structured extraction$0.0060$0.024
Agentic coding session, input mostly cached$0.074$0.298
Cost per benchmark task (AA Index)$1.27$4.00
Output tokens per task (AA Index)68k41k
Of which reasoning49k26k
Time per task (AA Index)24 min14 min
Output speed48 tok/s49 tok/s
Time to first answer134.5 s202.7 s
Time to a complete answer, last measured 13 Nov 2026429.6 s–

Per token is not per task: an average task of Artificial Analysis' Intelligence Index cost $1.27 on GPT-6.1 Sol (max) and $4.00 on GPT-6.2 Astra (max), and GPT-6.1 Sol (max) wrote 1.7× as many output tokens for the same work, which a per-token price hides.

Each model is compared at its best-ranked reasoning-effort setting. GPT-6.1 Sol (max) scores 65.9 and takes 135 s to its first answer token; GPT-6.1 Sol (xhigh) scores 62.9 in 59 s, GPT-6.1 Sol (high) scores 62.6 in 26 s. GPT-6.2 Astra (max) scores 68.9 and takes 203 s to its first answer token; GPT-6.2 Astra (xhigh) scores 67.1 in 67 s, GPT-6.2 Astra (medium) scores 65.6 in 4 s. 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

159 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-6.1 Sol (max)GPT-6.2 Astra (max)
BenchLeader IndexBenchLeader65.968.9
Agents & tools scoreBenchLeader63.166.4
Coding scoreBenchLeader70.176.1
Composite scoreBenchLeader65.569.5
Human preference scoreBenchLeader67.3–
Knowledge scoreBenchLeader70.374.2
Long context scoreBenchLeader66.2–
Maths scoreBenchLeader71.372.5
Multimodal scoreBenchLeader65.571.7
Reasoning scoreBenchLeader70.173.1
Blended price $/MBenchLeader$4.00$16.00
Output speedBenchLeader48 tok/s49 tok/s
Time to first answerBenchLeader134.5 s202.7 s
Context windowBenchLeader1,050,000 tokens1,050,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub95.4%96.8%
SimpleBenchSimpleBench82.9%†–
Arena Hard PromptsArena (LMArena)1507–
LiveBench 2.0 ReasoningLiveBench92.6%91.6%
LiveBench 1.0 ReasoningLiveBench92.6%–
Humanity's Last Exam (AA)Artificial Analysis52.9%–
HLE-Diamond (AA)Artificial Analysis58.8%72.2%
ARC-AGI-1ARC Prize96.5%98.5%
ARC-AGI-2ARC Prize94.2%95.8%
ARC-AGI-3ARC Prize96.2%99.8%
CritPtArtificial Analysis31.7%–
CritPt v2 hard subset (AA)Artificial Analysis69.5%71.5%
MysteryMechanismVals AI46.4%53.1%
HLE-Diamond (Scale)Scale AI SEAL53.6%†66.9%†
Arena ExpertArena (LMArena)1530–
Chess PuzzlesEpoch AI Benchmarking Hub68.0%78.0%
Mystery Game PuzzlesEpoch AI Benchmarking Hub80.0%95.0%
LMCAEpoch AI Benchmarking Hub62.8%–
DTBenchEpoch AI Benchmarking Hub96.5%–
Coding
SciCodeSciCode54.2%–
FrontierCodeCognition50.2%†–
Arena CodingArena (LMArena)1544–
Arena WebDevArena (LMArena)17441781
LiveBench 2.0 CodingLiveBench82.1%91.9%
LiveBench 1.0 CodingLiveBench80.4%–
Show all 159 benchmarks
MetricPublisherGPT-6.1 Sol (max)GPT-6.2 Astra (max)
Coding
IOIVals AI96.9%–
Code MigrationVals AI65.1%–
Code Migration v1.1Vals AI77.3%–
ProgramBenchVals AI3.0%–
SWE-bench MultimodalVals AI43.5%–
Vibe Code Bench v1.1Vals AI88.9%89.9%
Vibe Code Bench 1-100 v1.1Vals AI–55.2%
SWE Atlas v1.1Scale AI SEAL65.9%†60.5%†
SWE Atlas v1.1: Codebase QnAScale AI SEAL71.0%†67.7%†
SWE Atlas v1.1: Test WritingScale AI SEAL66.7%†62.2%†
SWE Atlas v1.1: RefactoringScale AI SEAL60.0%†51.4%†
SWE-Bench Pro V2Scale AI SEAL96.1%†–
MirrorCodeEpoch AI Benchmarking Hub61.1%67.8%
ALE-BenchEpoch AI Benchmarking Hub2743.8–
WeirdML v3WeirdML38.3%†–
Terminal-Bench 4.0 (AA)Artificial Analysis56.1%64.1%
Startrise FrontendStartrise78.6%†–
GameBench 2 OverallDuelLab GameBench 284.9†–
GameBench 2 by reasoning settingDuelLab GameBench 291.8†–
Design Arena WebsiteDesign Arena1311†1339†
FrontierCode 1.1 (AA)Artificial Analysis49.1%–
FrontierCode 2.0 (AA)Artificial Analysis40.6%45.4%
Agents & tools
Terminal-BenchTerminal-Bench60.9%62.1%
APEX-AgentsMercor60.0%–
Arena AgentArena (LMArena)11.315.2
LiveBench Long AgenticLiveBench64.7%70.8%
LiveBench BrowserLiveBench93.3%93.0%
LiveBench Agentic CodingLiveBench54.5%–
HiL-BenchScale AI SEAL62.3%†–
GDPval-AA v2.1, normalised scoreArtificial Analysis54.8%–
GDPval-AA v2.1Artificial Analysis1607–
Analyst Agent (AA)Artificial Analysis50.0%62.5%
CyberBenchVals AI39.3%–
Finance Agent v2Vals AI52.0%–
Finance Agent v2.1Vals AI61.8%60.8%
Harvey's Legal Agent Benchmark v1.1Vals AI6.3%16.3%
Harvey's Legal Agent Benchmark v1Vals AI5.4%–
Legal Research BenchVals AI38.5%42.8%
Robo-benchVals AI28.3%32.7%
SREBenchVals AI50.8%–
Tax Agent BenchVals AI21.8%21.8%
Terminal-Bench 4.0 (Vals)Vals AI55.0%63.1%
Terminal-Bench ScienceVals AI52.9%71.4%
Arena SearchArena (LMArena)1249–
Arena Agent Task CompletionArena (LMArena)11.612.4
PostTrainBenchEpoch AI Benchmarking Hub38.8%–
GBAEvalEpoch AI Benchmarking Hub81.9%†–
Vending-Bench 2Andon Labs12860.6†–
SpireClimbEpoch AI Benchmarking Hub23.8%41.3%
AutomationBenchArtificial Analysis64.9%77.3%
GDP.pdf (AA)Artificial Analysis31.0%29.8%
Harvey LAB-AA v1.1Harvey6.9%15.0%
EnterpriseOps-GymArtificial Analysis–29.9%
AA-Briefcase v1.1Artificial Analysis15571643
GDPxGDPx by SemiAnalysis8.4%–
Design Arena AgenticDesign Arena13591304
Terminal-Bench-Science 0.1 (AA)Artificial Analysis58.1%78.1%
CWE-Bench (AA)Artificial Analysis–65.0%
DeepSec-Bench (AA)Artificial Analysis–46.8%
CyberGym end-to-end (AA)Artificial Analysis–0.0%
MKA (AA)Artificial Analysis47.7%52.7%
FACTS SearchGoogle92.8%†–
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub93.7%92.6%
FrontierMath Tier 4Epoch AI Benchmarking Hub100.0%100.0%
OTIS Mock AIMEEpoch AI Benchmarking Hub100.0%100.0%
ProofBenchVals AI99.0%†–
LiveBench MathematicsLiveBench96.8%–
ArXivMathMathArena92.8%–
BrokenArXivMathArena83.1%–
Autoform-HardVals AI–87.5%
Arena MathsArena (LMArena)1500–
FrontierMath ErdősEpoch AI Benchmarking Hub4.4%10.3%
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub73.9%78.5%
LiveBench 2.0 Data AnalysisLiveBench82.7%84.1%
LiveBench ResearchLiveBench90.0%91.5%
LiveBench 1.0 Data AnalysisLiveBench82.7%–
AA-Omniscience v1.1Artificial Analysis42.050
PRBench FinanceScale AI SEAL46.2%†–
PRBench LegalScale AI SEAL47.9%†43.1%†
BioMysteryBenchVals AI79.6%80.7%
ClaimBenchVals AI100.0%–
Excel Modeling BenchmarkVals AI70.8%–
Excel Modeling Benchmark v1.1Vals AI71.1%76.0%
MedCodeVals AI48.8%–
MedScribeVals AI86.5%–
Public Benefits Bench v1.1Vals AI59.3%–
Access to Justice BenchmarkVals AI65.8%–
Financial Coaching BenchVals AI68.5%–
Immigration BenchVals AI67.8%–
DrugDiscoveryBenchScale AI SEAL–65.5%†
EBR-benchEpoch AI Benchmarking Hub54.3%92.4%
GDP.pdfEpoch AI Benchmarking Hub51.6%–
AA-Omniscience v1.1: accuracyArtificial Analysis61.8%68.6%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis48.1%40.9%
HealthBench Professional (AA)OpenAI65.1%–
FACTS ParametricGoogle83.8%†–
Instruction following
LiveBench LanguageLiveBench90.1%–
LiveBench Instruction FollowingLiveBench74.2%–
MultiChallengeScale AI SEAL82.0%†73.2%†
Arena Instruction FollowingArena (LMArena)1484–
Human preference
Arena TextArena (LMArena)1483–
EQ-Bench 4EQ-Bench1272†–
Arena Creative WritingArena (LMArena)1450–
Arena Multi-turnArena (LMArena)1487–
Arena Longer QueriesArena (LMArena)1490–
Multimodal
Arena VisionArena (LMArena)1286–
LiveBench DesignLiveBench79.6%82.8%
MMMU-ProArtificial Analysis86.0%89.5%
SAGEVals AI90.5%–
Humanity's Sixth SenseScale AI SEAL46.6%63.2%†
Blueprint-Bench 2Epoch AI Benchmarking Hub43.2%†–
Furniture AssemblyEpoch AI Benchmarking Hub80.0%89.2%
VIALS-AAHandshake AI59.0%65.8%
Code2Video-AAArtificial Analysis1369–
AA-3DArtificial Analysis11211318
FACTS MultimodalGoogle52.6%†–
Long context
AA-LCR v1.1Artificial Analysis83.0%–
MLCRArtificial Analysis38.3%39.4%
FACTS GroundingGoogle78.0%†–
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub168.6†174.2†
LiveBench 2.0LiveBench81.2%84.9%
LiveBench 1.0LiveBench81.6%–
AA Intelligence Index v5.0.1Artificial Analysis54.662.1
AA Intelligence Index v4.3.2Artificial Analysis51.8–
AA Progress Index v1Artificial Analysis348.5–
Vals IndexVals AI63.967.7
Vals Finance IndexVals AI55.8%57.3%
Vals RSI IndexVals AI26.6%29.0%
Vals Science IndexVals AI59.6%68.4%
Vals Router IndexVals AI63.9%†–
Vals Legal IndexVals AI68.7%69.7%
AA Cyber IndexArtificial Analysis–37.3
Safety & honesty
FORTRESSScale AI SEAL14.7%†–
DistressBenchScale AI SEAL74.2%†–

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

DuelLab GameBench 2, release gb2-20261114-ca113f475f5e (https://benchmarks.duellab.org/releases/gb2-20261114-ca113f475f5e/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • GPT-6.2 Astra (max): BenchLeader Index 67.8 → 68.9.
  • GPT-6.1 Sol (max): 2 score revisions.
  • GPT-6.2 Astra (max): 6 new results (VIALS-AA, Arena WebDev, Arena Agent…).
  • GPT-6.2 Astra (max): BenchLeader Index 67.2 → 67.8.
  • GPT-6.2 Astra (max): 2 new results (MKA (AA), Terminal-Bench).
  • GPT-6.1 Sol (max): BenchLeader Index 66.9 → 66.1.
  • GPT-6.1 Sol (max): 1 score revision.
  • GPT-6.1 Sol (max): 3 score revisions.
  • GPT-6.1 Sol (max): 8 new results (LiveBench 2.0, LiveBench Long Agentic, LiveBench 2.0 Reasoning…); 1 score revision.
  • GPT-6.1 Sol (max): 1 score revision.

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

GPT-6.1 Sol vs GPT-6.2 Astra: questions

Is GPT-6.2 Astra better than GPT-6.1 Sol?
GPT-6.2 Astra (max) leads on quality: 68.9 vs 65.9. The BenchLeader Index combines every independent quality benchmark; GPT-6.2 Astra (max) is ahead overall as of 19 Nov 2026, but check the category scores for your use.
GPT-6.1 Sol vs GPT-6.2 Astra: which is cheaper, and will that change?
GPT-6.1 Sol is cheaper today: $4.00 against $16.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is GPT-6.1 Sol or GPT-6.2 Astra better for coding?
GPT-6.2 Astra scores higher in coding (76 vs 70 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1744 for GPT-6.1 Sol against 1781 for GPT-6.2 Astra.
Which is faster, GPT-6.1 Sol or GPT-6.2 Astra?
GPT-6.2 Astra streams faster (49 vs 48 output tokens per second, Artificial Analysis 7-day median).
Is GPT-6.1 Sol or GPT-6.2 Astra better for agents and tool use?
GPT-6.2 Astra scores higher in agentic tasks (66 vs 63 on the category index, where 50 is average). On Terminal-Bench, the top shared agents & tools benchmark, it is 60.9% for GPT-6.1 Sol against 62.1% for GPT-6.2 Astra.
More questions (4)
Which is better for long documents?
Both accept 1,050,000 tokens of context.
Which has lower latency?
GPT-6.1 Sol answers first: 134.5 s against 202.7 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
Do independent benchmarks agree on GPT-6.1 Sol vs GPT-6.2 Astra?
Across the 61 benchmarks both have, GPT-6.1 Sol leads 7, GPT-6.2 Astra leads 52 and 2 are tied. Every overall index that rates both puts GPT-6.2 Astra ahead.
When was GPT-6.2 Astra released?
GPT-6.2 Astra was released on 17 Nov 2026.