Enter opens the highlighted result. Esc closes.

Gemini 3.8 Flash vs GPT-5.6 Terra

GPT-5.6 Terra leads Gemini 3.8 Flash on coding, 63.9 to 61.6 on the Vibe Coding Index, and Gemini 3.8 Flash costs 3.2x less per output token ($3.75 vs $12.00 per million). Updated October 22, 2026; 116 benchmarks have a result for both.

Replace Gemini 3.8 Flash Replace GPT-5.6 Terra
Gemini 3.8 FlashGoogle
Coding
61.6
Overall
69.0
Output $/M
$3.75
85
GPT-5.6 TerraOpenAI
Coding
63.9
Overall
68.1
Output $/M
$12.00
31
vs boards won of 116 shared

Gemini 3.8 Flash or GPT-5.6 Terra: which to pick

Pick GPT-5.6 Terra for coding: it scores 63.9 on the Vibe Coding Index against 61.6 for Gemini 3.8 Flash.

Pick Gemini 3.8 Flash for output tokens 3.2x cheaper ($3.75 vs $12.00 per million) and work measured by LMArena WebDev or LiveCodeBench (Vals AI), where it leads.

Model AGemini 3.8 FlashGoogle61.6/1006 independent benchmarks$2.74 per solved taskModel BGPT-5.6 TerraOpenAI63.9/1006 independent benchmarks$0.42 per solved task
Shared results
116
Gemini 3.8 Flash only
12
GPT-5.6 Terra only
11
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
Gemini 3.8 Flash 61.6 from 6 benchmarks, GPT-5.6 Terra 63.9 from 6 benchmarks.
Widest category
Math, 18.5 points apart on average over 6 shared benchmarks.
Widest single gap
FrontierMath, tier 4: Gemini 3.8 Flash 22.0% against GPT-5.6 Terra 70.7%, 48.8 points apart on a 0 to 100 scale.
Output price
Gemini 3.8 Flash $3.75 against GPT-5.6 Terra $12.00 per million output tokens, 3.2x apart.
Context window
Gemini 3.8 Flash takes 1.05M tokens, GPT-5.6 Terra 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
CategoryGemini 3.8 FlashBarsGPT-5.6 TerraTies
Coding1410
Intelligence02
Agentic work116
Reasoning141
Math24
Knowledge92
Web design81
General275
All shared boards8531
CodingLike-for-likeSix of these boards feed the Vibe Coding Index1410
Coding benchmarks for Gemini 3.8 Flash and GPT-5.6 Terra
BenchmarkGemini 3.8 FlashGPT-5.6 TerraLeader
LiveBench coding82.2%86.2%GPT-5.6 Terra
LMArena WebDev15851522Gemini 3.8 Flash
AA Coding Index76.376.7GPT-5.6 Terra
LiveCodeBench (Vals AI)89.5%85.9%Gemini 3.8 Flash
SWE-bench Verified (Vals AI)80.0%95.4%GPT-5.6 Terra
Terminal-Bench 4 (Vals AI)19.7%22.7%GPT-5.6 Terra
Vibe Code Bench (Vals AI)78.7%74.6%Gemini 3.8 Flash
Code migration (Vals AI)36.5%47.8%GPT-5.6 Terra
ProgramBench (Vals AI)1.0%0.5%Gemini 3.8 Flash
IOI (Vals AI)56.9%87.6%GPT-5.6 Terra
Terminal-Bench 2.1 (Vals AI)81.3%77.5%Gemini 3.8 Flash
SWE-bench Multimodal (Vals AI)40.8%Not measuredOnly Gemini 3.8 Flash measured
Vibe Code Bench 1-100 (Vals AI)18.8%14.8%Gemini 3.8 Flash
SkillsBench (Vals AI)58.0%58.9%GPT-5.6 Terra
CyberBench (Vals AI)43.8%72.1%GPT-5.6 Terra
SciCode56.6%55.0%Gemini 3.8 Flash
Terminal-Bench Hard (Artificial Analysis run)Not measured62.9%Only GPT-5.6 Terra measured
FrontierCode41.2%41.3%GPT-5.6 Terra
FrontierSWE19.6%Not measuredOnly Gemini 3.8 Flash measured
WeirdML84.8%78.3%Gemini 3.8 Flash
ALE-Bench1269.71951.4GPT-5.6 Terra
LMArena Text: coding15081485Gemini 3.8 Flash
LMArena Text: software and IT15071482Gemini 3.8 Flash
LMArena WebDev: React15771513Gemini 3.8 Flash
LMArena WebDev: HTML16161526Gemini 3.8 Flash
LMArena WebDev: from an image15931541Gemini 3.8 Flash
SciCode12.3%10.8%Gemini 3.8 Flash
Neither model has a result on the other 12 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for Gemini 3.8 Flash and GPT-5.6 Terra
BenchmarkGemini 3.8 FlashGPT-5.6 TerraLeader
AA Intelligence Index40.942.1GPT-5.6 Terra
Epoch Capabilities Index157.4160.1GPT-5.6 Terra
Agentic workLike-for-likePublic boards, shown as published, kept out of the index116
Agentic work benchmarks for Gemini 3.8 Flash and GPT-5.6 Terra
BenchmarkGemini 3.8 FlashGPT-5.6 TerraLeader
AA Agentic Index40.243.2GPT-5.6 Terra
Design Arena: agents building web apps12381120Gemini 3.8 Flash
Design Arena: agents building full-stack apps12441061Gemini 3.8 Flash
Design Arena: agents building mobile apps12361111Gemini 3.8 Flash
Design Arena: agents building Android apps10781062Gemini 3.8 Flash
Design Arena: agents building Godot games1273Not measuredOnly Gemini 3.8 Flash measured
Design Arena: agents building gamesNot measured1083Only GPT-5.6 Terra measured
Design Arena: HTML slidesNot measured1058Only GPT-5.6 Terra measured
Design Arena: PowerPoint slides11531105Gemini 3.8 Flash
Design Arena: spreadsheets1022Not measuredOnly Gemini 3.8 Flash measured
LiveBench research79.1%60.4%Gemini 3.8 Flash
LiveBench browser64.0%84.0%GPT-5.6 Terra
LiveBench long agentic30.1%48.4%GPT-5.6 Terra
SREBench (Vals AI)11.1%9.5%Gemini 3.8 Flash
Finance Agent (Vals AI)61.4%54.4%Gemini 3.8 Flash
Tax Agent Bench (Vals AI)32.1%30.0%Gemini 3.8 Flash
Legal Research Bench (Vals AI)38.9%41.4%GPT-5.6 Terra
Harvey's Legal Agent Benchmark (Vals AI)12.1%Not measuredOnly Gemini 3.8 Flash measured
Excel Modeling Benchmark (Vals AI)72.2%66.2%Gemini 3.8 Flash
BioMysteryBench (Vals AI)62.2%Not measuredOnly Gemini 3.8 Flash measured
Terminal-Bench-Science (Vals AI)8.6%10.0%GPT-5.6 Terra
Vals Science Index35.8%Not measuredOnly Gemini 3.8 Flash measured
Time Horizon Index (Vals AI)18.8%Not measuredOnly Gemini 3.8 Flash measured
CUA-bench (Vals AI)4.2%Not measuredOnly Gemini 3.8 Flash measured
τ²-Bench TelecomNot measured86.3%Only GPT-5.6 Terra measured
APEX-Agents64.3%58.2%Gemini 3.8 Flash
Remote Labor Index5.8%Not measuredOnly Gemini 3.8 Flash measured
BALROGNot measured53.2%Only GPT-5.6 Terra measured
Vending-Bench 2$5,094$7,343GPT-5.6 Terra
Neither model has a result on the other 12 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index141
Reasoning benchmarks for Gemini 3.8 Flash and GPT-5.6 Terra
BenchmarkGemini 3.8 FlashGPT-5.6 TerraLeader
GPQA Diamond94.4%90.9%Gemini 3.8 Flash
LiveBench reasoning89.3%90.6%GPT-5.6 Terra
MysteryMechanism (Vals AI)36.5%Not measuredOnly Gemini 3.8 Flash measured
MMMU Pro (Vals AI)89.1%86.5%Gemini 3.8 Flash
GPQA Diamond (Artificial Analysis run)95.3%92.5%Gemini 3.8 Flash
Humanity's Last Exam (Artificial Analysis run)47.8%42.9%Gemini 3.8 Flash
GPQA Diamond (Epoch AI run)95.4%93.3%Gemini 3.8 Flash
ARC-AGI98.5%96.5%Gemini 3.8 Flash
ARC-AGI-289.2%83.9%Gemini 3.8 Flash
SimpleBenchNot measured48.9%Only GPT-5.6 Terra measured
Chess puzzles58.0%47.0%Gemini 3.8 Flash
Mystery game puzzles47.0%35.0%Gemini 3.8 Flash
LMArena Text: hard prompts15091469Gemini 3.8 Flash
LMArena Text: expert15261496Gemini 3.8 Flash
LMArena Text: hard prompts in English15041465Gemini 3.8 Flash
LMArena Vision: diagrams13451295Gemini 3.8 Flash
LMArena Vision: homework13511312Gemini 3.8 Flash
Kagi LLM BenchmarkNot measured51.3%Only GPT-5.6 Terra measured
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index24
Math benchmarks for Gemini 3.8 Flash and GPT-5.6 Terra
BenchmarkGemini 3.8 FlashGPT-5.6 TerraLeader
OTIS Mock AIME98.9%99.7%GPT-5.6 Terra
FrontierMath, tiers 1 to 368.4%86.0%GPT-5.6 Terra
FrontierMath, tier 422.0%70.7%GPT-5.6 Terra
ProofBench48.0%74.0%GPT-5.6 Terra
LMArena Text: math15321465Gemini 3.8 Flash
LMArena Text: mathematical work15191478Gemini 3.8 Flash
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index92
Knowledge benchmarks for Gemini 3.8 Flash and GPT-5.6 Terra
BenchmarkGemini 3.8 FlashGPT-5.6 TerraLeader
MMLU Pro90.2%86.7%Gemini 3.8 Flash
LegalBench (Vals AI)87.0%85.1%Gemini 3.8 Flash
CorpFin (Vals AI)Not measured65.3%Only GPT-5.6 Terra measured
TaxEval (Vals AI)74.5%76.2%GPT-5.6 Terra
MortgageTax (Vals AI)65.3%67.3%GPT-5.6 Terra
MedCode (Vals AI)48.1%43.4%Gemini 3.8 Flash
SimpleQA Verified69.7%43.2%Gemini 3.8 Flash
LMArena Text: science15231452Gemini 3.8 Flash
LMArena Text: medicine and healthcare15081430Gemini 3.8 Flash
LMArena Text: legal and government14971461Gemini 3.8 Flash
LMArena Text: business and finance14791448Gemini 3.8 Flash
FACTS Benchmark Suite69.4%52.0%Gemini 3.8 Flash
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index81
Web design benchmarks for Gemini 3.8 Flash and GPT-5.6 Terra
BenchmarkGemini 3.8 FlashGPT-5.6 TerraLeader
Design Arena: websites13001268Gemini 3.8 Flash
Design Arena: UI components13161256Gemini 3.8 Flash
Design Arena: game dev13041229Gemini 3.8 Flash
Design Arena: data viz12701240Gemini 3.8 Flash
Design Arena: 3D12941134Gemini 3.8 Flash
Design Arena: all code categories13031241Gemini 3.8 Flash
Design Arena: SVG12961219Gemini 3.8 Flash
Design Arena: ASCII art12621177Gemini 3.8 Flash
LiveBench design72.7%76.0%GPT-5.6 Terra
GeneralLike-for-likePublic boards, shown as published, kept out of the index275
General benchmarks for Gemini 3.8 Flash and GPT-5.6 Terra
BenchmarkGemini 3.8 FlashGPT-5.6 TerraLeader
Vals Index56.0%Not measuredOnly Gemini 3.8 Flash measured
LiveBench data analysis54.0%79.3%GPT-5.6 Terra
Public Benefits Bench (Vals AI)65.3%62.4%Gemini 3.8 Flash
SAGE (Vals AI)35.1%47.0%GPT-5.6 Terra
MedScribe (Vals AI)84.5%82.9%Gemini 3.8 Flash
Vals Multimodal IndexNot measured65.1%Only GPT-5.6 Terra measured
IFBenchNot measured71.2%Only GPT-5.6 Terra measured
Long context reasoning (AA-LCR)84.0%83.0%Gemini 3.8 Flash
DTBench95.7%93.3%Gemini 3.8 Flash
LMCA52.9%54.9%GPT-5.6 Terra
Surface Evolver Bench76.9%83.8%GPT-5.6 Terra
Furniture assembly31.7%54.2%GPT-5.6 Terra
LMArena Text14991448Gemini 3.8 Flash
LMArena Text: instruction following14921456Gemini 3.8 Flash
LMArena Text: creative writing14881413Gemini 3.8 Flash
LMArena Vision13151271Gemini 3.8 Flash
LMArena Text: English14981449Gemini 3.8 Flash
LMArena Text: non-English14911441Gemini 3.8 Flash
LMArena Text: Chinese15531514Gemini 3.8 Flash
LMArena Text: German15011460Gemini 3.8 Flash
LMArena Text: Japanese15071454Gemini 3.8 Flash
LMArena Text: Korean14641427Gemini 3.8 Flash
LMArena Text: French14961466Gemini 3.8 Flash
LMArena Text: Spanish14871451Gemini 3.8 Flash
LMArena Text: Russian15141452Gemini 3.8 Flash
LMArena Text: Polish14971446Gemini 3.8 Flash
LMArena Text: multi-turn15031453Gemini 3.8 Flash
LMArena Text: longer queries15071454Gemini 3.8 Flash
LMArena Text: writing and language14941431Gemini 3.8 Flash
LMArena Text: entertainment and media14801418Gemini 3.8 Flash
LMArena Vision: English13181272Gemini 3.8 Flash
LMArena Vision: Chinese14161337Gemini 3.8 Flash
LMArena Vision: reading text13331278Gemini 3.8 Flash
LMArena Vision: humor13211236Gemini 3.8 Flash
LMArena Vision: creative writing12991248Gemini 3.8 Flash
LMArena DocumentNot measured1472Only GPT-5.6 Terra measured
Neither model has a result on the other 10 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemini 3.8 Flash and GPT-5.6 Terra
BenchmarkGemini 3.8 FlashGPT-5.6 TerraLeader
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

Gemini 3.8 Flash$0.0026Fits in one request
GPT-5.6 Terra$0.0080Fits in one request

Gemini 3.8 Flash costs less

Repository review

50K input + 3K output tokens

Gemini 3.8 Flash$0.049Fits in one request
GPT-5.6 Terra$0.136Fits in one request

Gemini 3.8 Flash costs less

Agent loop

200K input + 10K output tokens

Gemini 3.8 Flash$0.188Fits in one request
GPT-5.6 Terra$0.520Fits in one request

Gemini 3.8 Flash 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 Gemini 3.8 Flash and GPT-5.6 Terra
SpecificationGoogle Gemini 3.8 FlashOpenAI GPT-5.6 Terra
MakerGoogleOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedSeptember 2, 2026July 9, 2026
Context window1.05M1.05M
Input $/Mtok$0.75$2.00
Output $/Mtok$3.75$12.00
Coding index61.663.9
Independent measurements behind it66
Cost per solved task$2.74$0.42
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemini 3.8 Flash better than GPT-5.6 Terra for coding?

Pick GPT-5.6 Terra for coding: it scores 63.9 on the Vibe Coding Index against 61.6 for Gemini 3.8 Flash. Pick Gemini 3.8 Flash for output tokens 3.2x cheaper ($3.75 vs $12.00 per million) and work measured by LMArena WebDev or LiveCodeBench (Vals AI), where it leads.

Which is cheaper, Gemini 3.8 Flash or GPT-5.6 Terra?

Gemini 3.8 Flash costs $0.75 per million input tokens and $3.75 per million output tokens; GPT-5.6 Terra costs $2.00 and $12.00. Per solved SWE-bench Verified task, Gemini 3.8 Flash costs $2.74 and GPT-5.6 Terra $0.42.

Which has the bigger context window, Gemini 3.8 Flash or GPT-5.6 Terra?

GPT-5.6 Terra, with 1.05M tokens against 1.05M.

How many benchmarks is the Gemini 3.8 Flash vs GPT-5.6 Terra comparison based on?

116 benchmarks have a result for both models as of October 22, 2026. Gemini 3.8 Flash's coding index rests on 6 independent benchmarks and GPT-5.6 Terra'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/gemini-3-8-flash-vs-gpt-5-6-terra" title="Gemini 3.8 Flash vs GPT-5.6 Terra, 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/gemini-3-8-flash-vs-gpt-5-6-terra">LLMBenchmark.ai, Gemini 3.8 Flash vs GPT-5.6 Terra</a></p>