Enter opens the highlighted result. Esc closes.

GPT-6 Astra vs GPT-6 Sol

GPT-6 Astra leads GPT-6 Sol on coding, 85.1 to 76.0 on the Vibe Coding Index, and GPT-6 Sol costs 5.0x less per output token ($10.00 vs $50.00 per million). Updated October 22, 2026; 100 benchmarks have a result for both.

Replace GPT-6 Astra Replace GPT-6 Sol
GPT-6 AstraOpenAI
Coding
85.1
Overall
83.0
Output $/M
$50.00
93
GPT-6 SolOpenAI
Coding
76.0
Overall
74.6
Output $/M
$10.00
5
vs boards won of 100 shared, 2 tied

GPT-6 Astra or GPT-6 Sol: which to pick

Pick GPT-6 Astra for coding: it scores 85.1 on the Vibe Coding Index against 76.0 for GPT-6 Sol.

Pick GPT-6 Sol for output tokens 5.0x cheaper ($10.00 vs $50.00 per million) and work measured by CyberBench (Vals AI) or SciCode, where it leads.

Model AGPT-6 AstraOpenAI85.1/1005 independent benchmarksNo cost per solved taskModel BGPT-6 SolOpenAI76.0/1005 independent benchmarksNo cost per solved task
Shared results
100
GPT-6 Astra only
21
GPT-6 Sol only
1
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 Astra 85.1 from 5 benchmarks, GPT-6 Sol 76.0 from 5 benchmarks.
Widest category
Agentic work, 13.2 points apart on average over 17 shared benchmarks.
Widest single gap
CyberBench (Vals AI): GPT-6 Astra 41.1% against GPT-6 Sol 78.0%, 36.9 points apart on a 0 to 100 scale.
Output price
GPT-6 Astra $50.00 against GPT-6 Sol $10.00 per million output tokens, 5.0x apart.
Context window
Both take 1.05M tokens.

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 AstraBarsGPT-6 SolTies
Coding152
Intelligence20
Agentic work161
Reasoning140
Math501 tie
Knowledge70
Web design81
General2611 tie
All shared boards9352 ties
CodingLike-for-likeSix of these boards feed the Vibe Coding Index152
Coding benchmarks for GPT-6 Astra and GPT-6 Sol
BenchmarkGPT-6 AstraGPT-6 SolLeader
LiveBench coding93.7%93.1%GPT-6 Astra
LMArena WebDev17741683GPT-6 Astra
AA Coding Index76.9Not measuredOnly GPT-6 Astra measured
Terminal-Bench 4 (Vals AI)59.6%44.4%GPT-6 Astra
Vibe Code Bench (Vals AI)89.6%87.8%GPT-6 Astra
Code migration (Vals AI)67.7%57.2%GPT-6 Astra
ProgramBench (Vals AI)5.5%2.0%GPT-6 Astra
IOI (Vals AI)100.0%82.6%GPT-6 Astra
Terminal-Bench 2.1 (Vals AI)87.3%83.2%GPT-6 Astra
SWE-bench Multimodal (Vals AI)43.3%Not measuredOnly GPT-6 Astra measured
Vibe Code Bench 1-100 (Vals AI)27.6%Not measuredOnly GPT-6 Astra measured
CyberBench (Vals AI)41.1%78.0%GPT-6 Sol
SciCode56.5%57.6%GPT-6 Sol
FrontierCode53.3%49.3%GPT-6 Astra
FrontierSWE65.5%Not measuredOnly GPT-6 Astra measured
GSO-Bench79.4%Not measuredOnly GPT-6 Astra measured
WeirdML93.3%Not measuredOnly GPT-6 Astra measured
MirrorCode46.7%Not measuredOnly GPT-6 Astra measured
ALE-Bench2951.32462.0GPT-6 Astra
LMArena Text: coding14891448GPT-6 Astra
LMArena Text: software and IT14741433GPT-6 Astra
LMArena WebDev: React17981685GPT-6 Astra
LMArena WebDev: HTML17751685GPT-6 Astra
LMArena WebDev: from an image17261660GPT-6 Astra
SciCode12.3%Not measuredOnly GPT-6 Astra measured
Neither model has a result on the other 14 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for GPT-6 Astra and GPT-6 Sol
BenchmarkGPT-6 AstraGPT-6 SolLeader
AA Intelligence Index52.747.6GPT-6 Astra
Epoch Capabilities Index168.8163.7GPT-6 Astra
Agentic workLike-for-likePublic boards, shown as published, kept out of the index161
Agentic work benchmarks for GPT-6 Astra and GPT-6 Sol
BenchmarkGPT-6 AstraGPT-6 SolLeader
AA Agentic Index51.0Not measuredOnly GPT-6 Astra measured
Design Arena: agents building web apps13731231GPT-6 Astra
Design Arena: agents building full-stack apps13571238GPT-6 Astra
Design Arena: agents building mobile apps13121217GPT-6 Astra
Design Arena: agents building Godot games1348Not measuredOnly GPT-6 Astra measured
Design Arena: PowerPoint slides1219Not measuredOnly GPT-6 Astra measured
Design Arena: spreadsheets1365Not measuredOnly GPT-6 Astra measured
LiveBench research85.2%69.3%GPT-6 Astra
LiveBench browser84.0%92.0%GPT-6 Sol
LiveBench long agentic82.4%66.9%GPT-6 Astra
SREBench (Vals AI)56.9%30.5%GPT-6 Astra
Finance Agent (Vals AI)53.5%49.0%GPT-6 Astra
Tax Agent Bench (Vals AI)20.7%15.0%GPT-6 Astra
Legal Research Bench (Vals AI)39.4%28.9%GPT-6 Astra
Harvey's Legal Agent Benchmark (Vals AI)7.1%Not measuredOnly GPT-6 Astra measured
Excel Modeling Benchmark (Vals AI)71.7%71.5%GPT-6 Astra
BioMysteryBench (Vals AI)79.3%74.8%GPT-6 Astra
Terminal-Bench-Science (Vals AI)62.9%30.0%GPT-6 Astra
Vals Science Index65.1%45.0%GPT-6 Astra
Time Horizon Index (Vals AI)90.5%Not measuredOnly GPT-6 Astra measured
CUA-bench (Vals AI)19.2%Not measuredOnly GPT-6 Astra measured
Robo-bench (Vals AI)23.3%Not measuredOnly GPT-6 Astra measured
APEX-Agents64.7%54.3%GPT-6 Astra
Remote Labor Index20.8%Not measuredOnly GPT-6 Astra measured
BALROG68.3%Not measuredOnly GPT-6 Astra measured
SpireClimb25.7%7.5%GPT-6 Astra
Vending-Bench 2$15,515$14,428GPT-6 Astra
Neither model has a result on the other 14 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index140
Reasoning benchmarks for GPT-6 Astra and GPT-6 Sol
BenchmarkGPT-6 AstraGPT-6 SolLeader
LiveBench reasoning92.7%88.7%GPT-6 Astra
MysteryMechanism (Vals AI)53.1%30.2%GPT-6 Astra
GPQA Diamond (Artificial Analysis run)96.3%Not measuredOnly GPT-6 Astra measured
Humanity's Last Exam (Artificial Analysis run)54.7%47.9%GPT-6 Astra
GPQA Diamond (Epoch AI run)95.8%94.3%GPT-6 Astra
ARC-AGI98.5%95.5%GPT-6 Astra
ARC-AGI-295.0%89.6%GPT-6 Astra
Chess puzzles74.0%56.0%GPT-6 Astra
Mystery game puzzles84.0%56.0%GPT-6 Astra
EBR-bench76.2%53.3%GPT-6 Astra
LMArena Text: hard prompts14631422GPT-6 Astra
LMArena Text: expert14891437GPT-6 Astra
LMArena Text: hard prompts in English14641420GPT-6 Astra
LMArena Vision: diagrams12971270GPT-6 Astra
LMArena Vision: homework13371286GPT-6 Astra
Neither model has a result on the other 7 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index50
Math benchmarks for GPT-6 Astra and GPT-6 Sol
BenchmarkGPT-6 AstraGPT-6 SolLeader
OTIS Mock AIME100.0%100.0%Tie
FrontierMath, tiers 1 to 393.7%89.8%GPT-6 Astra
FrontierMath, tier 497.6%90.0%GPT-6 Astra
ProofBench99.0%83.0%GPT-6 Astra
LMArena Text: math14651407GPT-6 Astra
LMArena Text: mathematical work14841424GPT-6 Astra
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index70
Knowledge benchmarks for GPT-6 Astra and GPT-6 Sol
BenchmarkGPT-6 AstraGPT-6 SolLeader
MedCode (Vals AI)48.5%47.1%GPT-6 Astra
SimpleQA Verified75.6%60.7%GPT-6 Astra
LMArena Text: science14511389GPT-6 Astra
LMArena Text: medicine and healthcare14251366GPT-6 Astra
LMArena Text: legal and government14621396GPT-6 Astra
LMArena Text: business and finance14391380GPT-6 Astra
FACTS Benchmark Suite75.7%72.5%GPT-6 Astra
Neither model has a result on the other 11 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index81
Web design benchmarks for GPT-6 Astra and GPT-6 Sol
BenchmarkGPT-6 AstraGPT-6 SolLeader
Design Arena: websites13361292GPT-6 Astra
Design Arena: UI components13931290GPT-6 Astra
Design Arena: game dev14781311GPT-6 Astra
Design Arena: data viz13351269GPT-6 Astra
Design Arena: 3D15001314GPT-6 Astra
Design Arena: all code categories13791300GPT-6 Astra
Design Arena: SVG14311315GPT-6 Astra
Design Arena: ASCII art13531222GPT-6 Astra
LiveBench design82.0%82.2%GPT-6 Sol
GeneralLike-for-likePublic boards, shown as published, kept out of the index261
General benchmarks for GPT-6 Astra and GPT-6 Sol
BenchmarkGPT-6 AstraGPT-6 SolLeader
Vals Index65.0%Not measuredOnly GPT-6 Astra measured
LiveBench data analysis83.0%81.2%GPT-6 Astra
Public Benefits Bench (Vals AI)Not measured56.6%Only GPT-6 Sol measured
SAGE (Vals AI)46.4%44.8%GPT-6 Astra
MedScribe (Vals AI)87.9%82.0%GPT-6 Astra
Long context reasoning (AA-LCR)80.7%83.7%GPT-6 Sol
DTBench97.3%97.3%Tie
LMCA64.4%59.1%GPT-6 Astra
Furniture assembly80.0%58.3%GPT-6 Astra
LMArena Text14411398GPT-6 Astra
LMArena Text: instruction following14551413GPT-6 Astra
LMArena Text: creative writing14201378GPT-6 Astra
LMArena Vision12801240GPT-6 Astra
LMArena Text: English14471396GPT-6 Astra
LMArena Text: non-English14321388GPT-6 Astra
LMArena Text: Chinese14891412GPT-6 Astra
LMArena Text: German14591379GPT-6 Astra
LMArena Text: Japanese14051377GPT-6 Astra
LMArena Text: Korean14281336GPT-6 Astra
LMArena Text: French14321399GPT-6 Astra
LMArena Text: Spanish14221387GPT-6 Astra
LMArena Text: Russian14381402GPT-6 Astra
LMArena Text: Polish14221372GPT-6 Astra
LMArena Text: multi-turn14471414GPT-6 Astra
LMArena Text: longer queries14601416GPT-6 Astra
LMArena Text: writing and language14371389GPT-6 Astra
LMArena Text: entertainment and media14271373GPT-6 Astra
LMArena Vision: English12801229GPT-6 Astra
LMArena Vision: Chinese1361Not measuredOnly GPT-6 Astra measured
LMArena Vision: reading text12921258GPT-6 Astra
LMArena Vision: creative writing12861219GPT-6 Astra
Neither model has a result on the other 15 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-6 Astra and GPT-6 Sol
BenchmarkGPT-6 AstraGPT-6 SolLeader
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 Astra$0.035Fits in one request
GPT-6 Sol$0.0070Fits in one request

GPT-6 Sol costs less

Repository review

50K input + 3K output tokens

GPT-6 Astra$0.650Fits in one request
GPT-6 Sol$0.130Fits in one request

GPT-6 Sol costs less

Agent loop

200K input + 10K output tokens

GPT-6 Astra$2.50Fits in one request
GPT-6 Sol$0.500Fits in one request

GPT-6 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 Astra and GPT-6 Sol
SpecificationOpenAI GPT-6 AstraOpenAI GPT-6 Sol
MakerOpenAIOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedSeptember 4, 2026September 22, 2026
Context window1.05M1.05M
Input $/Mtok$10.00$2.00
Output $/Mtok$50.00$10.00
Coding index85.176.0
Independent measurements behind it55
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-6 Astra better than GPT-6 Sol for coding?

Pick GPT-6 Astra for coding: it scores 85.1 on the Vibe Coding Index against 76.0 for GPT-6 Sol. Pick GPT-6 Sol for output tokens 5.0x cheaper ($10.00 vs $50.00 per million) and work measured by CyberBench (Vals AI) or SciCode, where it leads.

Which is cheaper, GPT-6 Astra or GPT-6 Sol?

GPT-6 Astra costs $10.00 per million input tokens and $50.00 per million output tokens; GPT-6 Sol costs $2.00 and $10.00. Cost per solved task is known for neither of them.

Which has the bigger context window, GPT-6 Astra or GPT-6 Sol?

Neither: both take 1.05M tokens.

How many benchmarks is the GPT-6 Astra vs GPT-6 Sol comparison based on?

100 benchmarks have a result for both models as of October 22, 2026. GPT-6 Astra's coding index rests on 5 independent benchmarks and GPT-6 Sol's on 5. 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-astra-vs-gpt-6-sol" title="GPT-6 Astra vs GPT-6 Sol, 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-astra-vs-gpt-6-sol">LLMBenchmark.ai, GPT-6 Astra vs GPT-6 Sol</a></p>