Enter opens the highlighted result. Esc closes.

Claude Sonnet 5 vs GPT-5.6 Sol

GPT-5.6 Sol leads Claude Sonnet 5 on coding, 73.5 to 57.7 on the Vibe Coding Index, and Claude Sonnet 5 costs 2.0x less per output token ($10.00 vs $20.00 per million). Updated October 21, 2026; 119 benchmarks have a result for both.

Replace Claude Sonnet 5 Replace GPT-5.6 Sol
Claude Sonnet 5Anthropic
Coding
57.7
Overall
63.2
Output $/M
$10.00
13
GPT-5.6 SolOpenAI
Coding
73.5
Overall
74.7
Output $/M
$20.00
105
vs boards won of 119 shared, 1 tied

Claude Sonnet 5 or GPT-5.6 Sol: which to pick

Pick GPT-5.6 Sol for coding: it scores 73.5 on the Vibe Coding Index against 57.7 for Claude Sonnet 5.

Pick Claude Sonnet 5 for output tokens 2.0x cheaper ($10.00 vs $20.00 per million) and work measured by Vibe Code Bench (Vals AI) or LiveBench research, where it leads.

Model AClaude Sonnet 5Anthropic57.7/1006 independent benchmarks$1.87 per solved taskModel BGPT-5.6 SolOpenAI73.5/1006 independent benchmarks$1.20 per solved task
Shared results
119
Claude Sonnet 5 only
2
GPT-5.6 Sol only
20
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
Claude Sonnet 5 57.7 from 6 benchmarks, GPT-5.6 Sol 73.5 from 6 benchmarks.
Widest category
Math, 15.3 points apart on average over 6 shared benchmarks.
Widest single gap
FrontierMath, tier 4: Claude Sonnet 5 29.3% against GPT-5.6 Sol 82.9%, 53.7 points apart on a 0 to 100 scale.
Output price
Claude Sonnet 5 $10.00 against GPT-5.6 Sol $20.00 per million output tokens, 2.0x apart.
Context window
Claude Sonnet 5 takes 1M tokens, GPT-5.6 Sol 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
CategoryClaude Sonnet 5BarsGPT-5.6 SolTies
Coding124
Intelligence02
Agentic work2161 tie
Reasoning113
Math06
Knowledge65
Web design18
General231
All shared boards131051 tie
CodingLike-for-likeSix of these boards feed the Vibe Coding Index124
Coding benchmarks for Claude Sonnet 5 and GPT-5.6 Sol
BenchmarkClaude Sonnet 5GPT-5.6 SolLeader
LiveBench coding74.7%89.7%GPT-5.6 Sol
LMArena WebDev15401616GPT-5.6 Sol
AA Coding Index71.577.4GPT-5.6 Sol
LiveCodeBench (Vals AI)82.4%82.6%GPT-5.6 Sol
SWE-bench Verified (Vals AI)79.6%96.2%GPT-5.6 Sol
Terminal-Bench 4 (Vals AI)10.6%37.9%GPT-5.6 Sol
Vibe Code Bench (Vals AI)81.3%80.5%Claude Sonnet 5
Code migration (Vals AI)44.4%52.9%GPT-5.6 Sol
ProgramBench (Vals AI)0.0%1.5%GPT-5.6 Sol
IOI (Vals AI)45.0%91.2%GPT-5.6 Sol
Terminal-Bench 2.1 (Vals AI)74.5%85.8%GPT-5.6 Sol
Vibe Code Bench 1-100 (Vals AI)13.8%20.0%GPT-5.6 Sol
SkillsBench (Vals AI)46.5%54.1%GPT-5.6 Sol
CyberBench (Vals AI)70.3%76.3%GPT-5.6 Sol
SciCode54.3%57.8%GPT-5.6 Sol
Terminal-Bench Hard (Artificial Analysis run)Not measured65.9%Only GPT-5.6 Sol measured
FrontierCode42.7%47.5%GPT-5.6 Sol
FrontierSWENot measured32.2%Only GPT-5.6 Sol measured
GSO-Bench37.3%76.5%GPT-5.6 Sol
WeirdML68.8%88.8%GPT-5.6 Sol
MirrorCodeNot measured20.0%Only GPT-5.6 Sol measured
ALE-Bench1463.12176.9GPT-5.6 Sol
LMArena Text: coding14831496GPT-5.6 Sol
LMArena Text: software and IT14741489GPT-5.6 Sol
LMArena WebDev: React15411620GPT-5.6 Sol
LMArena WebDev: HTML15271618GPT-5.6 Sol
LMArena WebDev: from an image15401609GPT-5.6 Sol
SciCode10.7%10.8%GPT-5.6 Sol
Neither model has a result on the other 11 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for Claude Sonnet 5 and GPT-5.6 Sol
BenchmarkClaude Sonnet 5GPT-5.6 SolLeader
AA Intelligence Index38.247.0GPT-5.6 Sol
Epoch Capabilities Index156.9162.8GPT-5.6 Sol
Agentic workLike-for-likePublic boards, shown as published, kept out of the index216
Agentic work benchmarks for Claude Sonnet 5 and GPT-5.6 Sol
BenchmarkClaude Sonnet 5GPT-5.6 SolLeader
AA Agentic Index43.650.2GPT-5.6 Sol
Design Arena: agents building web apps12451245Tie
Design Arena: agents building full-stack apps12491261GPT-5.6 Sol
Design Arena: agents building mobile apps12151248GPT-5.6 Sol
Design Arena: agents building Android apps12251239GPT-5.6 Sol
Design Arena: agents building Godot games12251260GPT-5.6 Sol
Design Arena: agents building games12271267GPT-5.6 Sol
Design Arena: HTML slides12181286GPT-5.6 Sol
Design Arena: PowerPoint slides12161241GPT-5.6 Sol
Design Arena: spreadsheets1021Not measuredOnly Claude Sonnet 5 measured
LiveBench research82.8%54.5%Claude Sonnet 5
LiveBench browser72.0%88.0%GPT-5.6 Sol
LiveBench long agentic39.0%48.5%GPT-5.6 Sol
SREBench (Vals AI)0.4%30.5%GPT-5.6 Sol
Finance Agent (Vals AI)53.9%53.8%Claude Sonnet 5
Tax Agent Bench (Vals AI)29.4%31.1%GPT-5.6 Sol
Legal Research Bench (Vals AI)41.8%48.1%GPT-5.6 Sol
Excel Modeling Benchmark (Vals AI)66.3%72.3%GPT-5.6 Sol
BioMysteryBench (Vals AI)Not measured71.1%Only GPT-5.6 Sol measured
Terminal-Bench-Science (Vals AI)5.7%20.0%GPT-5.6 Sol
Vals Science IndexNot measured41.5%Only GPT-5.6 Sol measured
Time Horizon Index (Vals AI)Not measured23.8%Only GPT-5.6 Sol measured
CUA-bench (Vals AI)Not measured8.3%Only GPT-5.6 Sol measured
τ²-Bench TelecomNot measured85.1%Only GPT-5.6 Sol measured
APEX-Agents54.5%Not measuredOnly Claude Sonnet 5 measured
OSWorld 2.0Not measured27.3%Only GPT-5.6 Sol measured
BALROGNot measured60.0%Only GPT-5.6 Sol measured
PostTrainBenchNot measured36.2%Only GPT-5.6 Sol measured
Vending-Bench 2$6,378$9,619GPT-5.6 Sol
LMArena SearchNot measured1261Only GPT-5.6 Sol measured
Neither model has a result on the other 11 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index113
Reasoning benchmarks for Claude Sonnet 5 and GPT-5.6 Sol
BenchmarkClaude Sonnet 5GPT-5.6 SolLeader
GPQA Diamond88.9%95.2%GPT-5.6 Sol
LiveBench reasoning88.7%91.7%GPT-5.6 Sol
MysteryMechanism (Vals AI)Not measured33.3%Only GPT-5.6 Sol measured
MMMU Pro (Vals AI)83.0%88.8%GPT-5.6 Sol
GPQA Diamond (Artificial Analysis run)91.1%94.1%GPT-5.6 Sol
Humanity's Last Exam (Artificial Analysis run)41.3%49.5%GPT-5.6 Sol
GPQA Diamond (Epoch AI run)90.5%93.5%GPT-5.6 Sol
ARC-AGINot measured97.5%Only GPT-5.6 Sol measured
ARC-AGI-2Not measured92.5%Only GPT-5.6 Sol measured
SimpleBench60.6%64.8%GPT-5.6 Sol
Chess puzzles45.0%59.0%GPT-5.6 Sol
Mystery game puzzles35.0%58.0%GPT-5.6 Sol
EBR-benchNot measured44.8%Only GPT-5.6 Sol measured
LMArena Text: hard prompts14611483GPT-5.6 Sol
LMArena Text: expert14891517GPT-5.6 Sol
LMArena Text: hard prompts in English14641477GPT-5.6 Sol
LMArena Vision: diagrams12931305GPT-5.6 Sol
LMArena Vision: homework13111305Claude Sonnet 5
Kagi LLM BenchmarkNot measured67.0%Only GPT-5.6 Sol measured
Neither model has a result on the other 3 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index06
Math benchmarks for Claude Sonnet 5 and GPT-5.6 Sol
BenchmarkClaude Sonnet 5GPT-5.6 SolLeader
OTIS Mock AIME94.7%100.0%GPT-5.6 Sol
FrontierMath, tiers 1 to 365.6%89.1%GPT-5.6 Sol
FrontierMath, tier 429.3%82.9%GPT-5.6 Sol
ProofBench77.0%83.0%GPT-5.6 Sol
LMArena Text: math14671473GPT-5.6 Sol
LMArena Text: mathematical work14811496GPT-5.6 Sol
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index65
Knowledge benchmarks for Claude Sonnet 5 and GPT-5.6 Sol
BenchmarkClaude Sonnet 5GPT-5.6 SolLeader
MMLU Pro87.5%89.1%GPT-5.6 Sol
LegalBench (Vals AI)83.9%87.0%GPT-5.6 Sol
CorpFin (Vals AI)67.0%64.4%Claude Sonnet 5
TaxEval (Vals AI)75.6%74.8%Claude Sonnet 5
MortgageTax (Vals AI)70.0%67.3%Claude Sonnet 5
MedCode (Vals AI)47.5%44.0%Claude Sonnet 5
SimpleQA Verified33.7%69.7%GPT-5.6 Sol
LMArena Text: science14601455Claude Sonnet 5
LMArena Text: medicine and healthcare14421431Claude Sonnet 5
LMArena Text: legal and government14541476GPT-5.6 Sol
LMArena Text: business and finance14361457GPT-5.6 Sol
FACTS Benchmark SuiteNot measured52.4%Only GPT-5.6 Sol measured
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index18
Web design benchmarks for Claude Sonnet 5 and GPT-5.6 Sol
BenchmarkClaude Sonnet 5GPT-5.6 SolLeader
Design Arena: websites12741321GPT-5.6 Sol
Design Arena: UI components12801336GPT-5.6 Sol
Design Arena: game dev12781368GPT-5.6 Sol
Design Arena: data viz12511322GPT-5.6 Sol
Design Arena: 3D12611386GPT-5.6 Sol
Design Arena: all code categories12761341GPT-5.6 Sol
Design Arena: SVG11931316GPT-5.6 Sol
Design Arena: ASCII art12071304GPT-5.6 Sol
LiveBench design82.0%81.6%Claude Sonnet 5
GeneralLike-for-likePublic boards, shown as published, kept out of the index231
General benchmarks for Claude Sonnet 5 and GPT-5.6 Sol
BenchmarkClaude Sonnet 5GPT-5.6 SolLeader
LiveBench data analysis71.7%79.8%GPT-5.6 Sol
Public Benefits Bench (Vals AI)66.0%66.5%GPT-5.6 Sol
SAGE (Vals AI)48.9%52.6%GPT-5.6 Sol
MedScribe (Vals AI)76.0%85.2%GPT-5.6 Sol
Vals Multimodal Index68.8%72.6%GPT-5.6 Sol
IFBenchNot measured72.7%Only GPT-5.6 Sol measured
Long context reasoning (AA-LCR)82.0%84.0%GPT-5.6 Sol
DTBench92.5%95.7%GPT-5.6 Sol
LMCA50.0%58.5%GPT-5.6 Sol
Surface Evolver Bench60.0%93.1%GPT-5.6 Sol
Furniture assemblyNot measured56.7%Only GPT-5.6 Sol measured
LMArena Text14421457GPT-5.6 Sol
LMArena Text: instruction following14521483GPT-5.6 Sol
LMArena Text: creative writing14161449GPT-5.6 Sol
LMArena Vision12731281GPT-5.6 Sol
LMArena Text: English14501457GPT-5.6 Sol
LMArena Text: non-English14311451GPT-5.6 Sol
LMArena Text: Chinese14781525GPT-5.6 Sol
LMArena Text: German14421476GPT-5.6 Sol
LMArena Text: Japanese14221472GPT-5.6 Sol
LMArena Text: Korean14121442GPT-5.6 Sol
LMArena Text: French14631472GPT-5.6 Sol
LMArena Text: Spanish14381443GPT-5.6 Sol
LMArena Text: Russian14501471GPT-5.6 Sol
LMArena Text: Polish14251459GPT-5.6 Sol
LMArena Text: multi-turn14531461GPT-5.6 Sol
LMArena Text: longer queries14621481GPT-5.6 Sol
LMArena Text: writing and language14361462GPT-5.6 Sol
LMArena Text: entertainment and media14111450GPT-5.6 Sol
LMArena Vision: English12751280GPT-5.6 Sol
LMArena Vision: Chinese13311313Claude Sonnet 5
LMArena Vision: reading text12881290GPT-5.6 Sol
LMArena Vision: humor12611246Claude Sonnet 5
LMArena Vision: creative writing12461283GPT-5.6 Sol
LMArena Document14661483GPT-5.6 Sol
Neither model has a result on the other 11 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Sonnet 5 and GPT-5.6 Sol
BenchmarkClaude Sonnet 5GPT-5.6 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

Claude Sonnet 5$0.0070Fits in one request
GPT-5.6 Sol$0.014Fits in one request

Claude Sonnet 5 costs less

Repository review

50K input + 3K output tokens

Claude Sonnet 5$0.130Fits in one request
GPT-5.6 Sol$0.260Fits in one request

Claude Sonnet 5 costs less

Agent loop

200K input + 10K output tokens

Claude Sonnet 5$0.500Fits in one request
GPT-5.6 Sol$1.00Fits in one request

Claude Sonnet 5 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 Claude Sonnet 5 and GPT-5.6 Sol
SpecificationAnthropic Claude Sonnet 5OpenAI GPT-5.6 Sol
MakerAnthropicOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedJune 30, 2026July 9, 2026
Context window1M1.05M
Input $/Mtok$2.00$4.00
Output $/Mtok$10.00$20.00
Coding index57.773.5
Independent measurements behind it66
Cost per solved task$1.87$1.20
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Sonnet 5 better than GPT-5.6 Sol for coding?

Pick GPT-5.6 Sol for coding: it scores 73.5 on the Vibe Coding Index against 57.7 for Claude Sonnet 5. Pick Claude Sonnet 5 for output tokens 2.0x cheaper ($10.00 vs $20.00 per million) and work measured by Vibe Code Bench (Vals AI) or LiveBench research, where it leads.

Which is cheaper, Claude Sonnet 5 or GPT-5.6 Sol?

Claude Sonnet 5 costs $2.00 per million input tokens and $10.00 per million output tokens; GPT-5.6 Sol costs $4.00 and $20.00. Per solved SWE-bench Verified task, Claude Sonnet 5 costs $1.87 and GPT-5.6 Sol $1.20.

Which has the bigger context window, Claude Sonnet 5 or GPT-5.6 Sol?

GPT-5.6 Sol, with 1.05M tokens against 1M.

How many benchmarks is the Claude Sonnet 5 vs GPT-5.6 Sol comparison based on?

119 benchmarks have a result for both models as of October 21, 2026. Claude Sonnet 5's coding index rests on 6 independent benchmarks and GPT-5.6 Sol's on 6. 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/claude-sonnet-5-vs-gpt-5-6-sol" title="Claude Sonnet 5 vs GPT-5.6 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/claude-sonnet-5-vs-gpt-5-6-sol">LLMBenchmark.ai, Claude Sonnet 5 vs GPT-5.6 Sol</a></p>