BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases4

Gemini 3.6 Flash vs GPT-5.6 Terra

Full profiles: Gemini 3.6 Flash (Google) and GPT-5.6 Terra (OpenAI): every benchmark, price and rank.

Updated · Data from 11 sources

As of 2 Nov 2026, Gemini 3.6 Flash and GPT-5.6 Terra are neck and neck on quality (61.3 vs 61.1 on the BenchLeader Index, inside the ±5.8 margin), and Gemini 3.6 Flash is 3.0× cheaper at $1.50 vs $4.50 per 1M tokens blended. Choose Gemini 3.6 Flash for knowledge (64 vs 49), reasoning (64 vs 61), and multimodal (65 vs 63) and fast first answers (16.4 s vs 148.7 s), where it also streams ~1.5× faster; choose GPT-5.6 Terra for maths (69 vs 59), coding (65 vs 57), and long context (66 vs 65). Gemini 3.6 Flash's current price runs until 31 Dec 2026, then $3.00.

Key facts (as of 2 Nov 2026)

  • Gemini 3.6 Flash scores 61.3 on the BenchLeader Index and GPT-5.6 Terra scores 61.1 (BenchLeader, 2 Nov 2026).
  • Gemini 3.6 Flash costs $1.50 per 1M tokens blended until 31 Dec 2026, then $3.00 (https://ai.google.dev/gemini-api/docs/pricing, checked 20 Oct 2026).
  • GPT-5.6 Terra costs $4.50 per 1M tokens blended (list price, 2 Nov 2026).
  • Gemini 3.6 Flash streams ~1.5× faster than GPT-5.6 Terra (Artificial Analysis, 7-day median, 2 Nov 2026).
  • Of the 50 benchmarks both models have, GPT-5.6 Terra leads 39 (BenchLeader, 2 Nov 2026).
Gemini 3.6 Flash benchmarks high Google GPT-5.6 Terra benchmarks max OpenAIAdd a third model

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-5.6 Terra65 vs 57 category score; SciCode 55.0% vs 53.4%High
Agents & tool useEither59 vs 59 category score; τ³-Banking (AA) 29.9% vs 40.2%Either
MathsGPT-5.6 Terra69 vs 59 category score; FrontierMath Tiers 1–3 86.0% vs 59.0%High
ReasoningGemini 3.6 Flash64 vs 61 category score; GPQA Diamond 94.1% vs 93.3%Medium
KnowledgeGemini 3.6 Flash64 vs 49 category score; SimpleQA Verified 66.2% vs 43.2%High
Long documentsGPT-5.6 Terra66 vs 65 category score; AA-LCR v1.1 83.0% vs 80.0%Medium
Images and other mediaGemini 3.6 Flash65 vs 63 category score; MMMU-Pro 83.2% vs 80.7%Medium
Chat quality (human preference)–Only Gemini 3.6 Flash has an Arena rating (67 category score)Not comparable
Lowest costGemini 3.6 Flash$1.50 vs $4.50 per 1M blended; agentic coding session $0.060 vs $0.168; from 1 Jan 2027: $3.00 vs $4.50–
Fastest first answerGemini 3.6 Flash16.4 s (Gemini 3.6 Flash (high)) vs 148.7 s (GPT-5.6 Terra (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

  • Gemini 3.6 Flash (high) and GPT-5.6 Terra (max) are neck and neck on quality (61.3 vs 61.1, within the ±5.8 margin).
  • Gemini 3.6 Flash (high) is stronger in agents & tools, human preference, knowledge, multimodal, reasoning.
  • GPT-5.6 Terra (max) is stronger in coding, long context, maths, instruction following.
  • Gemini 3.6 Flash (high) is 3.0× cheaper ($1.50 vs $4.50 per 1M blended) until 31 Dec 2026, then $3.00 vs $4.50 from 1 Jan 2027.
  • Gemini 3.6 Flash (high) streams 1.5× faster (197 vs 130 tokens per second).

At a glance

SpecGemini 3.6 FlashGPT-5.6 Terra
DeveloperGoogleOpenAI
Released21 Jul 20269 Jul 2026
Context window (tokens)1,048,5761,050,000
Compared atGemini 3.6 Flash (high)GPT-5.6 Terra (max)
Open weightsNoNo
Price in / out / cached input per 1M$0.750 / $3.75 / $0.075$2.00 / $12.00 / $0.200
Announced price change$1.50 / $7.50 / $0.150 from 1 Jan 2027None announced

Limits, availability and announced prices from the vendors' own pages: ai.google.dev pricing (checked 20 Oct 2026). Everything else is from the day's data.

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.

CategoryGemini 3.6 Flash (high)GPT-5.6 Terra (max)Leader
Reasoning64.061.4Gemini 3.6 Flash
Coding56.964.9GPT-5.6 Terra
Agents & tools59.358.8level
Maths59.468.6GPT-5.6 Terra
Knowledge63.848.6Gemini 3.6 Flash
Instruction following–69.9not comparable
Human preference67.4–not comparable
Multimodal65.263.3Gemini 3.6 Flash
Long context64.866.4GPT-5.6 Terra

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 & toolsMathsKnowledgeLong contextMultimodalGemini 3.6 Flash, Reasoning: 64.0Gemini 3.6 Flash, Coding: 56.9Gemini 3.6 Flash, Agents & tools: 59.3Gemini 3.6 Flash, Maths: 59.4Gemini 3.6 Flash, Knowledge: 63.8Gemini 3.6 Flash, Long context: 64.8Gemini 3.6 Flash, Multimodal: 65.2GPT-5.6 Terra, Reasoning: 61.4GPT-5.6 Terra, Coding: 64.9GPT-5.6 Terra, Agents & tools: 58.8GPT-5.6 Terra, Maths: 68.6GPT-5.6 Terra, Knowledge: 48.6GPT-5.6 Terra, Long context: 66.4GPT-5.6 Terra, Multimodal: 63.3

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

ModelIndex

Where sources disagree

Across the 50 benchmarks both have, Gemini 3.6 Flash leads 11, GPT-5.6 Terra leads 39. Every overall index that rates both puts Gemini 3.6 Flash ahead.

Overall indexGemini 3.6 Flash (high)GPT-5.6 Terra (max)Leader
BenchLeader Index61.361.1neck and neck
Arena Text14831467†Gemini 3.6 Flash

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.

MeasureGemini 3.6 Flash (high)GPT-5.6 Terra (max)
Blended price per 1M tokens, until 31 Dec 2026$1.50$4.50
Input / output per 1M$0.750 / $3.75$2.00 / $12.00
Blended price per 1M from 1 Jan 2027$3.00$4.50
Input / output per 1M from 1 Jan 2027$1.50 / $7.50$2.00 / $12.00
Cached input per 1M$0.075$0.200
Price per task: chat reply$0.0014$0.0044
Price per task: summarize a 30-page report$0.011$0.031
Price per task: code edit$0.010$0.030
Price per task: agentic coding session$0.060$0.168
Price per task: structured extraction$0.0023$0.0064
Agentic coding session, input mostly cached$0.030$0.087
Output speed197 tok/s130 tok/s
Time to first answer16.4 s148.7 s
Time to a complete answer18.9 s152.5 s

Each model is compared at its best-ranked reasoning-effort setting. Gemini 3.6 Flash (high) scores 61.3 and takes 16 s to its first answer token; Gemini 3.6 Flash scores 49.5 in 1 s. GPT-5.6 Terra (max) scores 61.1 and takes 149 s to its first answer token; GPT-5.6 Terra (xhigh) scores 60.3 in 22 s, GPT-5.6 Terra (high) scores 57.0 in 5 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

122 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.

MetricPublisherGemini 3.6 Flash (high)GPT-5.6 Terra (max)
BenchLeader IndexBenchLeader61.361.1
Agents & tools scoreBenchLeader59.358.8
Coding scoreBenchLeader56.964.9
Human preference scoreBenchLeader67.4–
Knowledge scoreBenchLeader63.848.6
Long context scoreBenchLeader64.866.4
Maths scoreBenchLeader59.468.6
Multimodal scoreBenchLeader65.263.3
Reasoning scoreBenchLeader64.061.4
Instruction following scoreBenchLeader–69.9
Blended price $/MBenchLeader$1.50$4.50
Output speedBenchLeader197 tok/s130 tok/s
Time to first answerBenchLeader16.4 s148.7 s
Context windowBenchLeader1,048,576 tokens1,050,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub94.1%93.3%
SimpleBenchSimpleBench–48.9%†
Arena Hard PromptsArena (LMArena)15011490†
LiveBench 2.0 ReasoningLiveBench–90.6%†
LiveBench 1.0 ReasoningLiveBench85.2%90.6%
GPQA Diamond (AA)Artificial Analysis92.8%92.5%
Humanity's Last Exam (AA)Artificial Analysis40.8%42.9%
HLE-Diamond (AA)Artificial Analysis–28.8%
GPQA Diamond (Vals)Vals AI93.4%90.9%†
Kagi LLM BenchmarkKagi LLM Benchmark–51.3%†
ARC-AGI-1ARC Prize91.2%96.5%
ARC-AGI-2ARC Prize60.4%83.9%
ARC-AGI-3ARC Prize–0.8%
CritPtArtificial Analysis10.6%30.0%
Chess PuzzlesEpoch AI Benchmarking Hub40.0%47.0%
Mystery Game PuzzlesEpoch AI Benchmarking Hub30.0%35.0%
LMCAEpoch AI Benchmarking Hub44.9%55.0%
DTBenchEpoch AI Benchmarking Hub95.5%93.3%
ForecastBenchForecastBench65.9%†–
Coding
SciCodeSciCode53.4%55.0%
WeirdMLWeirdML56.1%78.3%†
FrontierCodeCognition34.4%†41.3%†
Arena CodingArena (LMArena)15181517†
Arena WebDevArena (LMArena)15381520†
LiveBench 2.0 CodingLiveBench–86.2%†
Show all 122 benchmarks
MetricPublisherGemini 3.6 Flash (high)GPT-5.6 Terra (max)
Coding
LiveBench 1.0 CodingLiveBench77.9%78.3%
LiveCodeBenchVals AI88.1%85.9%†
IOIVals AI35.1%87.6%
SWE-bench (Vals)Vals AI79.6%95.4%
Code MigrationVals AI30.9%47.8%
ProgramBenchVals AI0.0%0.5%
Vibe Code Bench 1-100Vals AI–14.8%
Vibe Code Bench v1.1Vals AI64.0%74.6%
Surface Evolver BenchEpoch AI Benchmarking Hub–83.8%†
DeepSWE v1.1Epoch AI Benchmarking Hub46.7%69.6%
CursorBenchCursor–41.3%
ALE-BenchEpoch AI Benchmarking Hub715.51951.4
Terminal-Bench 4.0 (AA)Artificial Analysis7.1%35.4%
Terminal-Bench 2.1 (AA)Artificial Analysis77.5%88.0%
Startrise FrontendStartrise–66.9%†
GameBench 2 OverallDuelLab GameBench 252.5†47.2†
GameBench 2 by reasoning settingDuelLab GameBench 264.654.8†
Agents & tools
Terminal-BenchTerminal-Bench–21.5%
APEX-AgentsMercor46.9%†58.2%
Arena AgentArena (LMArena)–-1.4†
LiveBench Long AgenticLiveBench–33.1%†
LiveBench BrowserLiveBench–70.3%†
LiveBench Agentic CodingLiveBench43.4%55.0%
Terminal-Bench HardArtificial Analysis–57.6%
τ²-Bench Telecom (AA)Artificial Analysis–86.3%
Terminal-Bench 2.1 (Vals)Vals AI73.8%77.5%
GDPval-AA v2.1, normalised scoreArtificial Analysis–47.6%
τ³-Banking (AA)Artificial Analysis29.9%40.2%
ITBench SRE (AA)Artificial Analysis–51.0%
APEX-Agents (AA)Artificial Analysis–38.9%
CyberBenchVals AI45.4%72.1%
Finance Agent v2Vals AI56.3%54.4%
Harvey's Legal Agent Benchmark v1Vals AI3.3%0.8%
Legal Research BenchVals AI25.0%41.4%
SkillsBenchVals AI–58.9%
SREBenchVals AI–9.5%
Tax Agent BenchVals AI18.0%30.0%
Terminal-Bench 4.0 (Vals)Vals AI–22.7%
Terminal-Bench ScienceVals AI4.3%10.0%
BALROGBALROG–53.2%
Vending-Bench 2Andon Labs–7343.2†
AutomationBenchArtificial Analysis53.0%59.6%
GDP.pdf (AA)Artificial Analysis17.4%24.0%
EnterpriseOps-GymArtificial Analysis–38.5%
Terminal-Bench-Science 0.1 (AA)Artificial Analysis–13.8%
FACTS SearchGoogle68.2%†81.7%†
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub59.0%86.0%
FrontierMath Tier 4Epoch AI Benchmarking Hub21.9%70.7%
OTIS Mock AIMEEpoch AI Benchmarking Hub94.2%99.7%
ProofBenchVals AI36.0%†74.0%†
LiveBench MathematicsLiveBench86.4%94.9%
AIME 2026MathArena96.7%†–
HMMT February 2026MathArena89.4%†–
MathArena ApexMathArena26.0%†–
Arena MathsArena (LMArena)15041480†
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub66.2%43.2%
LiveBench 2.0 Data AnalysisLiveBench–79.3%†
LiveBench ResearchLiveBench–60.4%†
LiveBench 1.0 Data AnalysisLiveBench63.0%79.3%
AA-Omniscience v1.1Artificial Analysis–-9.3
MMLU-ProVals AI89.3%86.7%†
LegalBenchVals AI86.7%85.1%
CorpFinVals AI63.3%65.3%†
TaxEvalVals AI74.9%76.2%†
BioMysteryBenchVals AI58.5%–
Excel Modeling BenchmarkVals AI65.4%66.2%
MedCodeVals AI53.1%43.4%†
MedScribeVals AI79.7%82.9%†
Public Benefits Bench v1.1Vals AI56.6%62.4%
GDP.pdfEpoch AI Benchmarking Hub14.0%24.7%†
AA-Omniscience v1.1: accuracyArtificial Analysis–42.3%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis–10.6%
Instruction following
LiveBench LanguageLiveBench83.9%82.9%
LiveBench Instruction FollowingLiveBench75.4%64.6%
IFBenchArtificial Analysis–71.2%
Arena Instruction FollowingArena (LMArena)14731463†
Human preference
Arena TextArena (LMArena)14831467†
EQ-Bench 4EQ-Bench–1220†
Arena Creative WritingArena (LMArena)14711429†
Arena Multi-turnArena (LMArena)14881474†
Arena Longer QueriesArena (LMArena)14851468†
Multimodal
Arena VisionArena (LMArena)12991271†
LiveBench DesignLiveBench–76.0%†
MMMU-ProArtificial Analysis83.2%80.7%
MMMU-Pro (Vals)Vals AI88.4%86.5%
MortgageTaxVals AI67.6%67.3%†
SAGEVals AI–87.0%
Blueprint-Bench 2Epoch AI Benchmarking Hub31.2%†30.8%†
Long context
AA-LCR v1.1Artificial Analysis80.0%83.0%
Arena DocumentArena (LMArena)14531477†
MLCRArtificial Analysis10.0%27.2%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub154.7†160.1†
LiveBench 2.0LiveBench–67.8%†
LiveBench 1.0LiveBench73.6%77.9%
AA Intelligence Index v4.3.2Artificial Analysis34.042.1
AA Progress Index v1Artificial Analysis305.7323.9
Vals Multimodal IndexVals AI65.1%65.1%†

Data as of 2026-11-02. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261026-cb69502a12d9 (https://benchmarks.duellab.org/releases/gb2-20261026-cb69502a12d9/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • Gemini 3.6 Flash: list price moves to $1.50 / $7.50 per 1M input / output tokens (announced).
  • GPT-5.6 Terra (max): 1 new result (MirrorCode).
  • Gemini 3.6 Flash (high): BenchLeader Index 61.8 → 61.2.
  • GPT-5.6 Terra (max): BenchLeader Index 62.5 → 61.2.
  • Gemini 3.6 Flash (high): 1 new result (AA Progress Index v1).
  • GPT-5.6 Terra (max): 2 new results (AA Progress Index v1, HLE-Diamond (AA)); 4 score revisions.
  • Gemini 3.6 Flash (high): 1 new result (Harvey's Legal Agent Benchmark v1).
  • GPT-5.6 Terra (max): 1 new result (Harvey's Legal Agent Benchmark v1).
  • Gemini 3.6 Flash (high): 1 score revision.
  • GPT-5.6 Terra (max): 1 score revision.

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

Gemini 3.6 Flash vs GPT-5.6 Terra: questions

Is Gemini 3.6 Flash better than GPT-5.6 Terra?
Gemini 3.6 Flash (high) and GPT-5.6 Terra (max) are neck and neck on quality (61.3 vs 61.1, within the ±5.8 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 2 Nov 2026, so check the category scores for your use.
Gemini 3.6 Flash vs GPT-5.6 Terra: which is cheaper, and will that change?
Gemini 3.6 Flash is cheaper today: $1.50 against $4.50 per million tokens, blended at three input tokens per output token. Gemini 3.6 Flash's current price runs until 31 Dec 2026, then $3.00.
Is Gemini 3.6 Flash or GPT-5.6 Terra better for coding?
GPT-5.6 Terra scores higher in coding (65 vs 57 on the category index, where 50 is average). On SciCode, the top shared coding benchmark, it is 53.4% for Gemini 3.6 Flash against 55.0% for GPT-5.6 Terra.
Which is faster, Gemini 3.6 Flash or GPT-5.6 Terra?
Gemini 3.6 Flash streams ~1.5× faster (197 vs 130 output tokens per second, Artificial Analysis 7-day median).
Is Gemini 3.6 Flash or GPT-5.6 Terra better for agents and tool use?
They are level in agentic tasks (59 vs 59 on the category index). On τ³-Banking (AA), the top shared agents & tools benchmark, it is 29.9% for Gemini 3.6 Flash against 40.2% for GPT-5.6 Terra.
More questions (4)
Which is better for long documents?
GPT-5.6 Terra scores higher on long-context benchmarks (66 vs 65). Both take about 1.1M tokens: Gemini 3.6 Flash 1,048,576, GPT-5.6 Terra 1,050,000.
Which has lower latency?
Gemini 3.6 Flash answers first: 16.4 s against 148.7 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 Gemini 3.6 Flash vs GPT-5.6 Terra?
Across the 50 benchmarks both have, Gemini 3.6 Flash leads 11, GPT-5.6 Terra leads 39. Every overall index that rates both puts Gemini 3.6 Flash ahead.
When was Gemini 3.6 Flash released?
Gemini 3.6 Flash was released on 21 Jul 2026.