Head-to-head comparison
Gemini 3.1 Pro Preview vs. Gemini 3.5 Flash-Lite
Compare published benchmark results from matching versions and cohorts, API pricing, and technical specifications. Any documented test setup differences remain visible.
143
shared result series
24
distinct benchmarks
2026-08-16
latest retrieval
Change model selection
Specifications and cost
General specifications and standard API pricing
Undisclosed parameter counts and knowledge cutoffs stay marked as unavailable. The table shows direct standard token pricing, including documented context tiers. Batch, fast, priority, regional surcharges, tool calls, and cloud platform pricing are excluded.
| Attribute | Gemini 3.1 Pro Preview | Gemini 3.5 Flash-Lite |
|---|---|---|
| Provider | ||
| Released | Feb 19, 2026 | Jul 21, 2026 |
| Availability | Preview | Active |
| Model type | Proprietary | Proprietary |
| Parameters | Not published | Not published |
| Architecture | Not published | Not published |
| Context window | 1,048,576 tokens | 1,048,576 tokens |
| Knowledge cutoff | Not published | Not published |
| Technical sources | Model | Model |
| API pricing per 1M tokens | ||
| API input | $2 ≤200K / $4 >200K | $0.3 |
| API output | $12 ≤200K / $18 >200K | $2.5 |
| Cache read | $0.2 ≤200K / $0.4 >200K | $0.03 |
| Cache write (5 min.) | Not listed | Not listed |
| Cache write (1 hr.) | Not listed | Not listed |
| Price verified | 07/30/2026Source | 07/30/2026Source |
Gemini 3.1 Pro PreviewGemini 3.5 Flash-Lite
API inputUSD per 1M tokens, base tier
Gemini 3.1 Pro Preview$2
Gemini 3.5 Flash-Lite$0.3Lower price
API outputUSD per 1M tokens, base tier
Gemini 3.1 Pro Preview$12
Gemini 3.5 Flash-Lite$2.5Lower price
Cache readUSD per 1M tokens, base tier
Gemini 3.1 Pro Preview$0.2
Gemini 3.5 Flash-Lite$0.03Lower price
Performance
Shared benchmarks
Only published results with the same benchmark version, task, metric, and comparison cohort are paired. Different reasoning levels, agents, harnesses, or output limits appear directly in the table.
143 shared result series
Gemini 3.1 Pro PreviewGemini 3.5 Flash-Lite
Arena Text, Style ControlOverall. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview1,486.245 arena pointsWinner
Gemini 3.5 Flash-Lite1,457.988 arena points
Code MigrationOverall. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview17.313 %Winner
Gemini 3.5 Flash-Lite6.148 %
CorpFin v2Overall. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview64.491 %Winner
Gemini 3.5 Flash-Lite60.684 %
CyberBenchOverall. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview36.441 %
Gemini 3.5 Flash-Lite72.316 %Winner
EMBOverall. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview52.616 %Winner
Gemini 3.5 Flash-Lite43.231 %
Finance Agent (v2)Overall. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview42.982 %
Gemini 3.5 Flash-Lite47.442 %Winner
GPQA DiamondOverall. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview95.454 %Winner
Gemini 3.5 Flash-Lite83.838 %
Harvey's Legal Agent BenchmarkOverall · Task fully resolved. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview0 %Tie
Gemini 3.5 Flash-Lite0 %Tie
Legal Research BenchOverall · All-pass. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview20.673 %Winner
Gemini 3.5 Flash-Lite13.942 %
LegalBenchOverall. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview87.398 %Winner
Gemini 3.5 Flash-Lite84.073 %
LiveCodeBench v6Overall. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview88.485 %Winner
Gemini 3.5 Flash-Lite79.014 %
MedCodeOverall. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview59.062 %Winner
Gemini 3.5 Flash-Lite43.489 %
| Benchmark and task | Gemini 3.1 Pro Preview | Gemini 3.5 Flash-Lite | Comparability | Source |
|---|---|---|---|---|
| Arena Text, Style ControlOverall | 1,486.245 arena pointsWinner 95% confidence interval: ±3.337The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | 1,457.988 arena points 95% confidence interval: ±5.907The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | style-control-v1style-controlled Bradley-Terry ratingHigher is betterCohort: lmarena-text-style-control:2026-08-12:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Snapshot: gemini-3.1-pro-preview, Harness: Arena text, style control Gemini 3.5 Flash-Lite: Snapshot: gemini-3.5-flash-lite, Harness: Arena text, style controlGemini 3.1 Pro Preview: Style-controlled Arena rating from 94,989 anonymous pairwise votes.Gemini 3.5 Flash-Lite: Style-controlled Arena rating from 13,676 anonymous pairwise votes. | ArenaType: Official benchmark leaderboardLeaderboard updated: 2026-08-12Import ID: lmarena-eeedf9209373a7f9b324bfb839e7b02fedacd1e5-1611ed64f3f1Retrieved 2026-08-16 |
| Code MigrationOverall | 17.313 %Winner | 6.148 % | 1accuracyHigher is betterCohort: vals-code-migration:1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-13Import ID: vals-code-migration-2026-08-16-107652dc6bfcRetrieved 2026-08-16 |
| CorpFin v2Overall | 64.491 %Winner | 60.684 % | 2accuracyHigher is betterCohort: vals-corp-fin-v2:2:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-corp_fin_v2-2026-08-16-45bbf84f57abRetrieved 2026-08-16 |
| CyberBenchOverall | 36.441 % | 72.316 %Winner | 1accuracyHigher is betterCohort: vals-cyberbench:1:overallNo documented setup differenceWinner: Gemini 3.5 Flash-LiteGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-03Import ID: vals-cyber-2026-08-16-702acaee9aa9Retrieved 2026-08-16 |
| EMBOverall | 52.616 %Winner | 43.231 % | 1accuracyHigher is betterCohort: vals-emb:1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-13Import ID: vals-emb-2026-08-16-68d5d1d47a97Retrieved 2026-08-16 |
| Finance Agent (v2)Overall | 42.982 % | 47.442 %Winner | 2accuracyHigher is betterCohort: vals-finance-agent-v2:2:overallNo documented setup differenceWinner: Gemini 3.5 Flash-LiteGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-14Import ID: vals-fabv2-2026-08-16-8853da43085dRetrieved 2026-08-16 |
| GPQA DiamondOverall | 95.454 %Winner | 83.838 % | 1accuracyHigher is betterCohort: vals-gpqa:1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-gpqa-2026-08-16-7f8363eaa2dbRetrieved 2026-08-16 |
| Harvey's Legal Agent BenchmarkOverall · Task fully resolved | 0 %Tie | 0 %Tie | 1task resolution rateHigher is betterCohort: vals-legal-agent-benchmark:1:overallNo documented setup differenceTieGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-14Import ID: vals-hlab-2026-08-16-bae90566df2fRetrieved 2026-08-16 |
| Legal Research BenchOverall · All-pass | 20.673 %Winner | 13.942 % | 1all-pass rateHigher is betterCohort: vals-legal-research:1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-14Import ID: vals-legal_research-2026-08-16-af296c00ecaaRetrieved 2026-08-16 |
| LegalBenchOverall | 87.398 %Winner | 84.073 % | 1accuracyHigher is betterCohort: vals-legal-bench:1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-legal_bench-2026-08-16-d5b3b1b2b5c8Retrieved 2026-08-16 |
| LiveCodeBench v6Overall | 88.485 %Winner | 79.014 % | 1accuracyHigher is betterCohort: vals-lcb:1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-13Import ID: vals-lcb-2026-08-16-17b9c147080dRetrieved 2026-08-16 |
| MedCodeOverall | 59.062 %Winner | 43.489 % | 1accuracyHigher is betterCohort: vals-medcode:1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-medcode-2026-08-16-524f7f474eacRetrieved 2026-08-16 |
| MedScribeOverall | 76.114 %Winner | 70.886 % | 1accuracyHigher is betterCohort: vals-medscribe:1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-medscribe-2026-08-16-c385354a7e62Retrieved 2026-08-16 |
| MMLU-ProOverall | 90.987 %Winner | 85.836 % | 1accuracyHigher is betterCohort: vals-mmlu-pro:1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-mmlu_pro-2026-08-16-e14d5bb3a34cRetrieved 2026-08-16 |
| MMMU-ProOverall | 88.208 %Winner | 83.584 % | 1accuracyHigher is betterCohort: vals-mmmu:1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-mmmu-2026-08-16-37c9a8c7e21fRetrieved 2026-08-16 |
| MortgageTaxOverall | 69.396 %Winner | 68.68 % | 1accuracyHigher is betterCohort: vals-mortgage-tax:1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-mortgage_tax-2026-08-16-e78c985dbb1cRetrieved 2026-08-16 |
| ProgramBenchOverall | 0 %Tie | 0 %Tie | 1accuracyHigher is betterCohort: vals-programbench:1:overallNo documented setup differenceTieGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-13Import ID: vals-programbench-2026-08-16-08d9727662ddRetrieved 2026-08-16 |
| Public Benefits Bench v1.1Overall | 53.789 %Winner | 51.827 % | 1.1accuracyHigher is betterCohort: vals-snap:1.1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-public-benefits-bench-2026-08-16-7ce68123c502Retrieved 2026-08-16 |
| SAGEOverall | 48.677 %Winner | 47.303 % | 1accuracyHigher is betterCohort: vals-sage:1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-sage-2026-08-16-a416dd25affbRetrieved 2026-08-16 |
| SWE-bench VerifiedOverall | 78.8 %Winner | 75 % | 1accuracyHigher is betterCohort: vals-swebench:1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-swebench-2026-08-16-c133071b8167Retrieved 2026-08-16 |
| TaxEval v2Overall | 72.882 %Winner | 72.608 % | 2accuracyHigher is betterCohort: vals-tax-eval-v2:2:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-tax_eval_v2-2026-08-16-5bbd495dac53Retrieved 2026-08-16 |
| Terminal-Bench 2.1Overall | 70.787 %Winner | 50.187 % | 2.1accuracyHigher is betterCohort: vals-terminal-bench-2-1:2.1:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-13Import ID: vals-terminal-bench-2-1-2026-08-16-b3f38c18bea4Retrieved 2026-08-16 |
| Vals IndexOverall | 41.903 %Winner | 36.711 % | 2weighted index scoreHigher is betterCohort: vals-vals-index:2:overallNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: No further settings disclosed | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-14Import ID: vals-vals_index-2026-08-16-ab641d89d6edRetrieved 2026-08-16 |
| Vibe Code Bench v1.1Overall | 32.034 % | 37.161 %Winner | 1.1accuracyHigher is betterCohort: vals-vibe-code-bench:1.1:overallNo documented setup differenceWinner: Gemini 3.5 Flash-LiteGemini 3.1 Pro Preview: Harness: OpenHands Gemini 3.5 Flash-Lite: Harness: OpenHands | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-13Import ID: vals-vibe-code-2026-08-16-a73c9770bcc4Retrieved 2026-08-16 |
| Arena Text, Style ControlBusiness, management and finance | 1,476.932 arena pointsWinner 95% confidence interval: ±5.678The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | 1,459.06 arena points 95% confidence interval: ±12.079The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | style-control-v1style-controlled Bradley-Terry ratingHigher is betterCohort: lmarena-text-style-control:2026-08-12:industry_business_and_management_and_financial_operationsNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Snapshot: gemini-3.1-pro-preview, Harness: Arena text, style control Gemini 3.5 Flash-Lite: Snapshot: gemini-3.5-flash-lite, Harness: Arena text, style controlGemini 3.1 Pro Preview: Style-controlled Arena rating from 18,759 anonymous pairwise votes.Gemini 3.5 Flash-Lite: Style-controlled Arena rating from 2,561 anonymous pairwise votes. | ArenaType: Official benchmark leaderboardLeaderboard updated: 2026-08-12Import ID: lmarena-eeedf9209373a7f9b324bfb839e7b02fedacd1e5-1611ed64f3f1Retrieved 2026-08-16 |
| Code MigrationCLIReasoning: high | 10.833 %Winner | 5.875 % | 1accuracyHigher is betterCohort: vals-code-migration:1:cliNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-code-migration-2026-08-11-3a2d5ce718a3Retrieved 2026-08-11 |
| CorpFin v2Exact PagesReasoning: high | 60.839 %Winner | 57.459 % | 2accuracyHigher is betterCohort: vals-corp-fin-v2:2:exact_pagesNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-corp_fin_v2-2026-08-11-d9e58d191cd9Retrieved 2026-08-11 |
| CyberBenchPatch | 69.492 % | 77.966 %Higher measured value | 1accuracyHigher is betterCohort: vals-cyberbench:1:patchTest setup not documented as identicalNo direct winnerGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-03Import ID: vals-cyber-2026-08-11-1f20498a3dc1Retrieved 2026-08-11 |
| EMBCompsReasoning: high | 44.505 % | 46.417 %Winner | 1accuracyHigher is betterCohort: vals-emb:1:compsNo documented setup differenceWinner: Gemini 3.5 Flash-LiteGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-emb-2026-08-11-e3c855e236e4Retrieved 2026-08-11 |
| Finance Agent (v2)AdjustmentsReasoning: high | 38.361 % | 41.741 %Winner | 2accuracyHigher is betterCohort: vals-finance-agent-v2:2:adjustmentsNo documented setup differenceWinner: Gemini 3.5 Flash-LiteGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-fabv2-2026-08-11-3e5c93a1857cRetrieved 2026-08-11 |
| GPQA DiamondFew-Shot CoTReasoning: high | 95.96 %Higher measured value | 81.818 % | 1accuracyHigher is betterCohort: vals-gpqa:1:diamond_few_shot_cotTest setup not documented as identicalNo direct winnerGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,535 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-gpqa-2026-08-11-d3325290a404Retrieved 2026-08-11 |
| Legal Research BenchAdministrative / Regulatory · All-passReasoning: high | 24.242 %Winner | 19.697 % | 1all-pass rateHigher is betterCohort: vals-legal-research:1:administrative_regulatoryNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-legal_research-2026-08-11-326cf7aaca58Retrieved 2026-08-11 |
| LegalBenchConclusion TasksReasoning: high | 89.257 %Winner | 88.404 % | 1accuracyHigher is betterCohort: vals-legal-bench:1:conclusion_tasksNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-legal_bench-2026-08-11-836e5e44618eRetrieved 2026-08-11 |
| LiveCodeBench v6EasyReasoning: high | 99.068 %Winner | 98.758 % | 1accuracyHigher is betterCohort: vals-lcb:1:easyNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-lcb-2026-08-11-0342bcc364e4Retrieved 2026-08-11 |
| MMLU-ProBiologyReasoning: high | 95.258 %Higher measured value | 91.911 % | 1accuracyHigher is betterCohort: vals-mmlu-pro:1:biologyTest setup not documented as identicalNo direct winnerGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,535 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-mmlu_pro-2026-08-11-b8e53828d746Retrieved 2026-08-11 |
| MortgageTaxNumerical ExtractionReasoning: high | 76.391 %Winner | 75.835 % | 1accuracyHigher is betterCohort: vals-mortgage-tax:1:numerical_extractionNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-mortgage_tax-2026-08-11-5ea6eef3e482Retrieved 2026-08-11 |
| ProgramBenchAlmost ResolvedReasoning: high | 1 %Winner | 0 % | 1accuracyHigher is betterCohort: vals-programbench:1:almostNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-programbench-2026-08-11-b6f8959fbdfeRetrieved 2026-08-11 |
| SAGEAutomataReasoning: high | 32.143 %Winner | 28.968 % | 1accuracyHigher is betterCohort: vals-sage:1:automataNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 30,000 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 30,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-sage-2026-08-11-f1f79f6d8af3Retrieved 2026-08-11 |
| SWE-bench Verified<15 min fixReasoning: high | 88.66 %Winner | 85.567 % | 1accuracyHigher is betterCohort: vals-swebench:1:<15 min fixNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-08Import ID: vals-swebench-2026-08-11-a6212cd857d8Retrieved 2026-08-11 |
| TaxEval v2CorrectnessReasoning: high | 60.749 % | 61.733 %Higher measured value | 2accuracyHigher is betterCohort: vals-tax-eval-v2:2:correctnessTest setup not documented as identicalNo direct winnerGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,535 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-tax_eval_v2-2026-08-11-d05a40bf446fRetrieved 2026-08-11 |
| Terminal-Bench 2.1EasyReasoning: high | 91.667 %Winner | 66.667 % | 2.1accuracyHigher is betterCohort: vals-terminal-bench-2-1:2.1:easyNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-terminal-bench-2-1-2026-08-11-8f0ebd72a609Retrieved 2026-08-11 |
| Vals IndexCorpFin v2Reasoning: high | 65.734 %Winner | 62.238 % | 1.2benchmark scoreHigher is betterCohort: vals-vals-index:1.2:corp_fin_v2No documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-vals_index-2026-08-11-e79d08d59249Retrieved 2026-08-11 |
| Arena Text, Style ControlChinese | 1,531.944 arena pointsWinner 95% confidence interval: ±9.483The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | 1,494.257 arena points 95% confidence interval: ±19.894The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | style-control-v1style-controlled Bradley-Terry ratingHigher is betterCohort: lmarena-text-style-control:2026-08-12:chineseNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Snapshot: gemini-3.1-pro-preview, Harness: Arena text, style control Gemini 3.5 Flash-Lite: Snapshot: gemini-3.5-flash-lite, Harness: Arena text, style controlGemini 3.1 Pro Preview: Style-controlled Arena rating from 5,308 anonymous pairwise votes.Gemini 3.5 Flash-Lite: Style-controlled Arena rating from 977 anonymous pairwise votes. | ArenaType: Official benchmark leaderboardLeaderboard updated: 2026-08-12Import ID: lmarena-eeedf9209373a7f9b324bfb839e7b02fedacd1e5-1611ed64f3f1Retrieved 2026-08-16 |
| Code MigrationCOBOLReasoning: high | 36.753 %Winner | 6.964 % | 1accuracyHigher is betterCohort: vals-code-migration:1:cobolNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-code-migration-2026-08-11-3a2d5ce718a3Retrieved 2026-08-11 |
| CorpFin v2Max Fitting ContextReasoning: high | 66.9 %Winner | 62.354 % | 2accuracyHigher is betterCohort: vals-corp-fin-v2:2:max_fitting_contextNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-corp_fin_v2-2026-08-11-d9e58d191cd9Retrieved 2026-08-11 |
| CyberBenchPoC | 3.39 % | 66.667 %Higher measured value | 1accuracyHigher is betterCohort: vals-cyberbench:1:pocTest setup not documented as identicalNo direct winnerGemini 3.1 Pro Preview: No further settings disclosed Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-03Import ID: vals-cyber-2026-08-11-1f20498a3dc1Retrieved 2026-08-11 |
| EMBDataroomReasoning: high | 79.17 %Winner | 57.687 % | 1accuracyHigher is betterCohort: vals-emb:1:dataroomNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-emb-2026-08-11-e3c855e236e4Retrieved 2026-08-11 |
| Finance Agent (v2)All-PassReasoning: high | 30.879 % | 35.265 %Winner | 2accuracyHigher is betterCohort: vals-finance-agent-v2:2:all_passNo documented setup differenceWinner: Gemini 3.5 Flash-LiteGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-fabv2-2026-08-11-3e5c93a1857cRetrieved 2026-08-11 |
| GPQA DiamondZero-Shot CoTReasoning: high | 94.949 %Higher measured value | 85.859 % | 1accuracyHigher is betterCohort: vals-gpqa:1:diamond_zero_shot_cotTest setup not documented as identicalNo direct winnerGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,535 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-gpqa-2026-08-11-d3325290a404Retrieved 2026-08-11 |
| Legal Research BenchAdministrative / Regulatory · Weighted partial creditReasoning: high | 79.215 %Winner | 71.96 % | 1weighted partial-credit scoreHigher is betterCohort: vals-legal-research:1:weighted_pass__administrative_regulatoryNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-legal_research-2026-08-11-326cf7aaca58Retrieved 2026-08-11 |
| LegalBenchInterpretation TasksReasoning: high | 83.823 %Winner | 80.332 % | 1accuracyHigher is betterCohort: vals-legal-bench:1:interpretation_tasksNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-legal_bench-2026-08-11-836e5e44618eRetrieved 2026-08-11 |
| LiveCodeBench v6HardReasoning: high | 76.571 %Winner | 53.429 % | 1accuracyHigher is betterCohort: vals-lcb:1:hardNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-lcb-2026-08-11-0342bcc364e4Retrieved 2026-08-11 |
| MMLU-ProBusinessReasoning: high | 93.409 %Higher measured value | 90.621 % | 1accuracyHigher is betterCohort: vals-mmlu-pro:1:businessTest setup not documented as identicalNo direct winnerGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,535 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-mmlu_pro-2026-08-11-b8e53828d746Retrieved 2026-08-11 |
| MortgageTaxSemantic ExtractionReasoning: high | 62.401 %Winner | 61.526 % | 1accuracyHigher is betterCohort: vals-mortgage-tax:1:semantic_extractionNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-mortgage_tax-2026-08-11-5ea6eef3e482Retrieved 2026-08-11 |
| ProgramBenchFully ResolvedReasoning: high | 0 %Tie | 0 %Tie | 1accuracyHigher is betterCohort: vals-programbench:1:strictNo documented setup differenceTieGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-programbench-2026-08-11-b6f8959fbdfeRetrieved 2026-08-11 |
| SAGECalculusReasoning: high | 78.395 %Winner | 60.494 % | 1accuracyHigher is betterCohort: vals-sage:1:calculusNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 30,000 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 30,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-sage-2026-08-11-f1f79f6d8af3Retrieved 2026-08-11 |
| SWE-bench Verified>4 hoursReasoning: high | 33.333 %Tie | 33.333 %Tie | 1accuracyHigher is betterCohort: vals-swebench:1:>4 hoursNo documented setup differenceTieGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-08Import ID: vals-swebench-2026-08-11-a6212cd857d8Retrieved 2026-08-11 |
| TaxEval v2Stepwise ReasoningReasoning: high | 85.016 %Higher measured value | 83.483 % | 2accuracyHigher is betterCohort: vals-tax-eval-v2:2:stepwise_reasoningTest setup not documented as identicalNo direct winnerGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,535 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-tax_eval_v2-2026-08-11-d05a40bf446fRetrieved 2026-08-11 |
| Terminal-Bench 2.1HardReasoning: high | 60 %Winner | 31.111 % | 2.1accuracyHigher is betterCohort: vals-terminal-bench-2-1:2.1:hardNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-terminal-bench-2-1-2026-08-11-8f0ebd72a609Retrieved 2026-08-11 |
| Vals IndexFinance Agent v2Reasoning: high | 42.968 % | 46.626 %Winner | 1.2benchmark scoreHigher is betterCohort: vals-vals-index:1.2:finance_agentNo documented setup differenceWinner: Gemini 3.5 Flash-LiteGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-vals_index-2026-08-11-e79d08d59249Retrieved 2026-08-11 |
| Arena Text, Style ControlCoding | 1,521.319 arena pointsWinner 95% confidence interval: ±5.155The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | 1,502.808 arena points 95% confidence interval: ±9.913The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | style-control-v1style-controlled Bradley-Terry ratingHigher is betterCohort: lmarena-text-style-control:2026-08-12:codingNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Snapshot: gemini-3.1-pro-preview, Harness: Arena text, style control Gemini 3.5 Flash-Lite: Snapshot: gemini-3.5-flash-lite, Harness: Arena text, style controlGemini 3.1 Pro Preview: Style-controlled Arena rating from 26,285 anonymous pairwise votes.Gemini 3.5 Flash-Lite: Style-controlled Arena rating from 3,938 anonymous pairwise votes. | ArenaType: Official benchmark leaderboardLeaderboard updated: 2026-08-12Import ID: lmarena-eeedf9209373a7f9b324bfb839e7b02fedacd1e5-1611ed64f3f1Retrieved 2026-08-16 |
| Code MigrationCode QualityReasoning: high | 2.538 / 5 | 2.652 / 5Winner | 1mean reviewer code-quality ratingHigher is betterCohort: vals-code-migration:1:code_qualityNo documented setup differenceWinner: Gemini 3.5 Flash-LiteGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-code-migration-2026-08-11-3a2d5ce718a3Retrieved 2026-08-11 |
| CorpFin v2Shared Max ContextReasoning: high | 65.734 %Winner | 62.238 % | 2accuracyHigher is betterCohort: vals-corp-fin-v2:2:shared_max_contextNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-corp_fin_v2-2026-08-11-d9e58d191cd9Retrieved 2026-08-11 |
| EMBDCFReasoning: high | 46.388 %Winner | 33.614 % | 1accuracyHigher is betterCohort: vals-emb:1:dcfNo documented setup differenceWinner: Gemini 3.1 Pro PreviewGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens Gemini 3.5 Flash-Lite: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-emb-2026-08-11-e3c855e236e4Retrieved 2026-08-11 |
The table shows 64 diversified result series from 143 shared results. This prevents one benchmark with many subtasks from crowding out every other test.
How to read this comparison
Winning one benchmark is not an overall verdict
Your workload, budget, and required context length matter most. A coding benchmark says little about visual understanding or agent performance.
Prices use official API rates per 1M tokens. Web subscriptions and cloud platform surcharges can differ.
Every result links to its measurement source and retrieval date. Labels distinguish vendor reports, official benchmark leaderboards, and independent evaluations. Disputed or archived results are excluded.
More matchups
Related LLM comparisons
Compare Gemini 3.1 Pro Preview and Gemini 3.5 Flash-Lite with other leading models using the same data and benchmark logic.
View the complete LLM comparison