Skip to main content

Head-to-head comparison

Gemini 3.1 Pro Preview vs. DeepSeek-V4-Pro

Compare published benchmark results from matching versions and cohorts, API pricing, and technical specifications. Any documented test setup differences remain visible.

shared result series
168
distinct benchmarks
23
latest retrieval
2026-09-04

Change model selection

Specifications

General specifications

Release date, availability, context window, architecture, and published parameter counts at a glance. Undisclosed details remain marked as unavailable.

Specifications for Gemini 3.1 Pro Preview and DeepSeek-V4-Pro
AttributeGemini 3.1 Pro PreviewDeepSeek-V4-Pro
ProviderGoogleDeepSeek
ReleasedFeb 19, 2026Apr 24, 2026
AvailabilityPreviewPreview
Model typeProprietaryOpen Weights
ParametersNot published1.6T, 49B active
ArchitectureNot publishedMixture of Experts
Context window1,048,576 tokens1,000,000 tokens
Knowledge cutoffNot publishedNot published
Technical sourcesModelModel, Context

Pricing

API pricing compared

The table compares official standard API rates per 1M tokens, including published cache pricing. Batch, fast, priority, regional surcharges, tool calls, and cloud platform pricing are excluded.

API pricing for Gemini 3.1 Pro Preview and DeepSeek-V4-Pro
Price typeGemini 3.1 Pro PreviewDeepSeek-V4-Pro
API input$2 ≤200K / $4 >200K$0.66
API output$12 ≤200K / $18 >200K$1.98
Cache read$0.2 ≤200K / $0.4 >200K$0.02
Cache write (5 min.)Not listedNot listed
Cache write (1 hr.)Not listedNot listed
Price verified07/30/2026 Source09/08/2026 Source
Gemini 3.1 Pro PreviewDeepSeek-V4-Pro
API inputUSD per 1M tokens, base tier
Gemini 3.1 Pro Preview$2
DeepSeek-V4-ProLower price$0.66
API outputUSD per 1M tokens, base tier
Gemini 3.1 Pro Preview$12
DeepSeek-V4-ProLower price$1.98
Cache readUSD per 1M tokens, base tier
Gemini 3.1 Pro Preview$0.2
DeepSeek-V4-ProLower price$0.022

Key to the marks:better valuebetter, but not conclusivetie

Pricing by provider

Direct developer pricing and provider endpoint tariffs routed through OpenRouter are listed separately. Prices are in USD per 1M tokens. Context-dependent direct tiers plus regional, Flex, Priority, and other OpenRouter tariffs remain individually identifiable.

Provider pricing for Gemini 3.1 Pro Preview and DeepSeek-V4-Pro
ModelProviderPurchase route and tariffInput per 1M tokensOutput per 1M tokensSource
Gemini 3.1 Pro PreviewGoogleDirect from developerUp to 200,000 context tokens$2$12Source07/30/2026
GoogleDirect from developerAbove 200,000 context tokens$4$18Source07/30/2026
GoogleVia OpenRoutergoogle-vertex/global$2$12Source09/08/2026
GoogleVia OpenRoutergoogle-vertex/global/flexLowest listed price: $1Lowest listed price: $6Source09/08/2026
GoogleVia OpenRoutergoogle-vertex/global/priority$3.6$21.6Source09/08/2026
Google AI StudioVia OpenRoutergoogle-ai-studio$2$12Source09/08/2026
Google AI StudioVia OpenRoutergoogle-ai-studio/flexLowest listed price: $1Lowest listed price: $6Source09/08/2026
Google AI StudioVia OpenRoutergoogle-ai-studio/priority$3.6$21.6Source09/08/2026
DeepSeek-V4-ProDeepSeekDirect from developerStandard APILowest listed price: $0.66Lowest listed price: $1.98Source09/08/2026
AlibabaVia OpenRouteralibaba$1.122$3.366Source09/08/2026
BaiduVia OpenRouterbaidu/fp8$1.1194$3.3581Source09/08/2026
BaseTenVia OpenRouterbaseten/fp4#1$1.32$3.96Source09/08/2026
BaseTenVia OpenRouterbaseten/fp4#2$1.32$3.96Source09/08/2026
CloudflareVia OpenRoutercloudflare$1.32$3.96Source09/08/2026
CoreWeaveVia OpenRoutercoreweave/fp8$1.31$3.96Source09/08/2026
DeepInfraVia OpenRouterdeepinfra/fp8$1.3$2.6Source09/08/2026
DeepSeekVia OpenRouterdeepseekLowest listed price: $0.66$1.98Source09/08/2026
DigitalOceanVia OpenRouterdigitalocean$1.32$3.96Source09/08/2026
FireworksVia OpenRouterfireworks$1.32$3.96Source09/08/2026
GMICloudVia OpenRoutergmicloud/fp8$1.056$3.168Source09/08/2026
IonstreamVia OpenRouterionstream/fp4$1.1$3.3Source09/08/2026
NextBitVia OpenRouternextbit/fp8$1.122$3.366Source09/08/2026
NovitaVia OpenRouternovita/fp8$0.99$2.97Source09/08/2026
ParasailVia OpenRouterparasail/fp8$1.32$3.96Source09/08/2026
PhalaVia OpenRouterphala$1.45$4.36Source09/08/2026
Sail ResearchVia OpenRoutersail-research/fp4$1.32$3.96Source09/08/2026
SiliconFlowVia OpenRoutersiliconflow/fp8$1.32$3.96Source09/08/2026
StreamLakeVia OpenRouterstreamlake$1.0494$3.1482Source09/08/2026
TogetherVia OpenRoutertogether$1.32$3.96Source09/08/2026

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.

Capability profile from matched benchmarks

Each axis averages directly matched benchmark families. A value of 100 means the stronger value within this pair, not a universal quality score.

255075100CodingReasoningMathematicsKnowledgeFinanceLegalAgentic tasksProductivity
Gemini 3.1 Pro Preview
DeepSeek-V4-Pro
Coding: 88.4 / 99.3 (6 benchmark families)
Reasoning: 100 / 96 (2 benchmark families)
Mathematics: 100 / 79.6 (2 benchmark families)
Knowledge: 100 / 97 (2 benchmark families)
Finance: 99.5 / 98.2 (5 benchmark families)
Legal: 72.4 / 97.6 (4 benchmark families)
Agentic tasks: 100 / 77.1 (1 benchmark family)
Productivity: 85.5 / 100 (1 benchmark family)

54 matched rows from 22 benchmark families were considered. The radar shows 8 of 12 comparable categories.

Gemini 3.1 Pro PreviewDeepSeek-V4-Pro
Code MigrationOverall. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview17.313 %
DeepSeek-V4-ProWinner26.201 %
CorpFin v2Overall. Higher is better. No documented setup difference
Gemini 3.1 Pro PreviewWinner64.491 %
DeepSeek-V4-Pro61.383 %
CyberBenchOverall. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview36.441 %
DeepSeek-V4-ProWinner68.644 %
EMBOverall. Higher is better. No documented setup difference
Gemini 3.1 Pro PreviewWinner52.616 %
DeepSeek-V4-Pro51.623 %
Finance Agent (v2)Overall. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview42.982 %
DeepSeek-V4-ProWinner44.083 %
GPQA DiamondOverall. Higher is better. No documented setup difference
Gemini 3.1 Pro PreviewWinner95.454 %
DeepSeek-V4-Pro89.394 %
Harvey's Legal Agent BenchmarkOverall · Task fully resolved. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview0 %
DeepSeek-V4-ProWinner3.75 %
Legal Research BenchOverall · All-pass. Higher is better. No documented setup difference
Gemini 3.1 Pro Preview20.673 %
DeepSeek-V4-ProWinner23.077 %
LegalBenchOverall. Higher is better. No documented setup difference
Gemini 3.1 Pro PreviewWinner87.398 %
DeepSeek-V4-Pro80.323 %
LiveCodeBench v6Overall. Higher is better. No documented setup difference
Gemini 3.1 Pro PreviewWinner88.485 %
DeepSeek-V4-Pro87.484 %
LMArena Text, Style ControlOverall. Higher is better. No documented setup difference
Gemini 3.1 Pro PreviewWinner1,486.746
DeepSeek-V4-Pro1,457.748
MedCodeOverall. Higher is better. No documented setup difference
Gemini 3.1 Pro PreviewWinner59.062 %
DeepSeek-V4-Pro40.455 %

Key to the marks:better valuebetter, but not conclusivetie

Benchmark scores for Gemini 3.1 Pro Preview and DeepSeek-V4-Pro
Benchmark and taskGemini 3.1 Pro PreviewDeepSeek-V4-ProSource
Code MigrationOverall
Test details
No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better
17.313 %
Winner26.201 %
Vals AI
Source details
Data as of: 09/03/2026
CorpFin v2Overall
Test details
No documented difference in the test setupVersion: 2Metric: accuracyScoring: Higher is better
Winner64.491 %
61.383 %
Vals AI
Source details
Data as of: 08/12/2026
CyberBenchOverall
Test details
No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better
36.441 %
Winner68.644 %
Vals AI
Source details
Data as of: 09/01/2026
EMBOverall
Test details
No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better
Winner52.616 %
51.623 %
Vals AI
Source details
Data as of: 09/03/2026
Finance Agent (v2)Overall
Test details
No documented difference in the test setupVersion: 2Metric: accuracyScoring: Higher is better
42.982 %
Winner44.083 %
Vals AI
Source details
Data as of: 09/03/2026
GPQA DiamondOverall
Test details
No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better
Winner95.454 %
89.394 %
Vals AI
Source details
Data as of: 09/01/2026
Harvey's Legal Agent BenchmarkOverall · Task fully resolved
Test details
No documented difference in the test setupVersion: 1Metric: task resolution rateScoring: Higher is better
0 %
Winner3.75 %
Vals AI
Source details
Data as of: 09/03/2026
Legal Research BenchOverall · All-pass
Test details
No documented difference in the test setupVersion: 1Metric: all-pass rateScoring: Higher is better
20.673 %
Winner23.077 %
Vals AI
Source details
Data as of: 09/03/2026
LegalBenchOverall
Test details
No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better
Winner87.398 %
80.323 %
Vals AI
Source details
Data as of: 09/01/2026
LiveCodeBench v6Overall
Test details
No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better
Winner88.485 %
87.484 %
Vals AI
Source details
Data as of: 09/01/2026
LMArena Text, Style ControlOverall
Test details
No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGemini 3.1 Pro Preview: Snapshot: gemini-3.1-pro-preview, Harness: Arena text, style control
DeepSeek-V4-Pro: Snapshot: deepseek-v4-pro, Harness: Arena text, style control
Gemini 3.1 Pro Preview: Style-controlled Arena rating from 102,763 anonymous pairwise votes.DeepSeek-V4-Pro: Style-controlled Arena rating from 54,243 anonymous pairwise votes.
Winner1,486.746
95% confidence interval: ±3.217The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
1,457.748
95% confidence interval: ±4.009The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena
Source details
Data as of: 09/01/2026
MedCodeOverall
Test details
No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better
Winner59.062 %
40.455 %
Vals AI
Source details
Data as of: 09/03/2026
MedScribeOverall
Test details
No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better
Winner76.114 %
75.144 %
Vals AI
Source details
Data as of: 09/03/2026
MMLU-ProOverall
Test details
No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better
Winner90.987 %
87.249 %
Vals AI
Source details
Data as of: 09/01/2026
ProgramBenchOverall
Test details
No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better
Tie0 %
Tie0 %
Vals AI
Source details
Data as of: 09/01/2026
ProofBenchOverall
Test details
No documented difference in the test setupVersion: 1.1Metric: accuracyScoring: Higher is better
Winner26 %
16 %
Vals AI
Source details
Data as of: 09/01/2026
Public Benefits Bench v1.1Overall
Test details
No documented difference in the test setupVersion: 1.1Metric: accuracyScoring: Higher is better
53.789 %
Winner62.923 %
Vals AI
Source details
Data as of: 09/01/2026
SWE-bench VerifiedOverall
Test details
No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better
Winner78.8 %
77.4 %
Vals AI
Source details
Data as of: 09/01/2026
TaxEval v2Overall
Test details
No documented difference in the test setupVersion: 2Metric: accuracyScoring: Higher is better
Winner72.882 %
72.077 %
Vals AI
Source details
Data as of: 09/01/2026
Terminal-Bench 2.0Overall
Test details
No documented difference in the test setupVersion: 2.0Metric: accuracyScoring: Higher is better
Winner67.416 %
56.18 %
Vals AI
Source details
Data as of: 06/04/2026
Terminal-Bench 2.1Overall
Test details
No documented difference in the test setupVersion: 2.1Metric: accuracyScoring: Higher is better
Winner70.787 %
50.187 %
Vals AI
Source details
Data as of: 09/03/2026
Vals IndexOverall
Test details
No documented difference in the test setupVersion: 2Metric: weighted index scoreScoring: Higher is better
41.903 %
Winner42.888 %
Vals AI
Source details
Data as of: 09/03/2026
Vibe Code Bench v1.1Overall
Test details
No documented difference in the test setupVersion: 1.1Metric: accuracyScoring: Higher is betterSettings: Harness: OpenHands
32.034 %
Winner49.931 %
Vals AI
Source details
Data as of: 09/03/2026
Code MigrationCLI
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
10.833 %
Higher measured value15.47 %
Vals AI
Source details
Data as of: 08/09/2026
CorpFin v2Exact Pages
Test details
Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 128,000 tokens
Higher measured value60.839 %
59.207 %
Vals AI
Source details
Data as of: 08/06/2026
CyberBenchPatch
Test details
No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better
69.492 %
Winner79.661 %
Vals AI
Source details
Data as of: 08/03/2026
EMBComps
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
44.505 %
Higher measured value57.434 %
Vals AI
Source details
Data as of: 08/06/2026
Finance Agent (v2)Adjustments
Test details
Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
38.361 %
Higher measured value41.586 %
Vals AI
Source details
Data as of: 08/06/2026
GPQA DiamondFew-Shot CoT
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,535 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 256,000 tokens
Higher measured value95.96 %
90.404 %
Vals AI
Source details
Data as of: 08/09/2026
Legal Research BenchAdministrative / Regulatory · All-pass
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
24.242 %
Higher measured value28.788 %
Vals AI
Source details
Data as of: 08/06/2026
LegalBenchConclusion Tasks
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 256,000 tokens
Higher measured value89.257 %
89.117 %
Vals AI
Source details
Data as of: 08/09/2026
LiveCodeBench v6Easy
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 128,000 tokens
Higher measured value99.068 %
98.758 %
Vals AI
Source details
Data as of: 08/09/2026
LMArena Text, Style ControlOverall
Test details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGemini 3.1 Pro Preview: Snapshot: gemini-3.1-pro-preview, Harness: Arena text, style control
DeepSeek-V4-Pro: Snapshot: deepseek-v4-pro-high-preview, Reasoning: high, Harness: Arena text, style control
Gemini 3.1 Pro Preview: Style-controlled Arena rating from 102,763 anonymous pairwise votes.DeepSeek-V4-Pro: Style-controlled Arena rating from 51,599 anonymous pairwise votes.
Higher measured value1,486.746
95% confidence interval: ±3.217The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
1,455.047
95% confidence interval: ±4.08The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena
Source details
Data as of: 09/01/2026
MMLU-ProBiology
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,535 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 256,000 tokens
Higher measured value95.258 %
92.887 %
Vals AI
Source details
Data as of: 08/09/2026
ProgramBenchAlmost Resolved
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
Higher measured value1 %
0 %
Vals AI
Source details
Data as of: 08/09/2026
SWE-bench Verified<15 min fix
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 256,000 tokens
Higher measured value88.66 %
87.629 %
Vals AI
Source details
Data as of: 08/08/2026
TaxEval v2Correctness
Test details
Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,535 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 128,000 tokens
Higher measured value60.749 %
60.589 %
Vals AI
Source details
Data as of: 08/09/2026
Terminal-Bench 2.0Easy
Test details
Test setup not documented as identicalNo direct winnerVersion: 2.0Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Output limit: 128,000 tokens
Higher measured value100 %
75 %
Vals AI
Source details
Data as of: 06/04/2026
Terminal-Bench 2.1Easy
Test details
Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Output limit: 128,000 tokens
Higher measured value91.667 %
66.667 %
Vals AI
Source details
Data as of: 08/06/2026
Vals IndexCorpFin v2
Test details
Test setup not documented as identicalNo direct winnerVersion: 1.2Metric: benchmark scoreScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 128,000 tokens
Higher measured value65.734 %
61.888 %
Vals AI
Source details
Data as of: 08/09/2026
Code MigrationCOBOL
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
36.753 %
Higher measured value58.393 %
Vals AI
Source details
Data as of: 08/09/2026
CorpFin v2Max Fitting Context
Test details
Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 128,000 tokens
Higher measured value66.9 %
63.054 %
Vals AI
Source details
Data as of: 08/06/2026
CyberBenchPoC
Test details
No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better
3.39 %
Winner57.627 %
Vals AI
Source details
Data as of: 08/03/2026
EMBDataroom
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
Higher measured value79.17 %
62.572 %
Vals AI
Source details
Data as of: 08/06/2026
Finance Agent (v2)All-Pass
Test details
Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
30.879 %
Higher measured value31.448 %
Vals AI
Source details
Data as of: 08/06/2026
GPQA DiamondZero-Shot CoT
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,535 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 256,000 tokens
Higher measured value94.949 %
88.384 %
Vals AI
Source details
Data as of: 08/09/2026
Legal Research BenchAdministrative / Regulatory · Weighted partial credit
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: weighted partial-credit scoreScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
79.215 %
Higher measured value83.372 %
Vals AI
Source details
Data as of: 08/06/2026
LegalBenchInterpretation Tasks
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 256,000 tokens
Higher measured value83.823 %
75.169 %
Vals AI
Source details
Data as of: 08/09/2026
LiveCodeBench v6Hard
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 128,000 tokens
Higher measured value76.571 %
72.571 %
Vals AI
Source details
Data as of: 08/09/2026
LMArena Text, Style ControlBusiness, management and finance
Test details
No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGemini 3.1 Pro Preview: Snapshot: gemini-3.1-pro-preview, Harness: Arena text, style control
DeepSeek-V4-Pro: Snapshot: deepseek-v4-pro, Harness: Arena text, style control
Gemini 3.1 Pro Preview: Style-controlled Arena rating from 20,283 anonymous pairwise votes.DeepSeek-V4-Pro: Style-controlled Arena rating from 10,742 anonymous pairwise votes.
Winner1,476.52
95% confidence interval: ±5.472The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
1,461.034
95% confidence interval: ±6.908The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena
Source details
Data as of: 09/01/2026
MMLU-ProBusiness
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,535 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 256,000 tokens
Higher measured value93.409 %
91.888 %
Vals AI
Source details
Data as of: 08/09/2026
ProgramBenchFully Resolved
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
No direct winner0 %
No direct winner0 %
Vals AI
Source details
Data as of: 08/09/2026
SWE-bench Verified>4 hours
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 256,000 tokens
Higher measured value33.333 %
0 %
Vals AI
Source details
Data as of: 08/08/2026
TaxEval v2Stepwise Reasoning
Test details
Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,535 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 128,000 tokens
Higher measured value85.016 %
83.565 %
Vals AI
Source details
Data as of: 08/09/2026
Terminal-Bench 2.0Hard
Test details
Test setup not documented as identicalNo direct winnerVersion: 2.0Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Output limit: 128,000 tokens
Higher measured value53.333 %
33.333 %
Vals AI
Source details
Data as of: 06/04/2026
Terminal-Bench 2.1Hard
Test details
Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Output limit: 128,000 tokens
Higher measured value60 %
31.111 %
Vals AI
Source details
Data as of: 08/06/2026
Vals IndexFinance Agent v2
Test details
Test setup not documented as identicalNo direct winnerVersion: 1.2Metric: benchmark scoreScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
42.968 %
Higher measured value44.081 %
Vals AI
Source details
Data as of: 08/09/2026
Code MigrationCode Quality
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: mean reviewer code-quality ratingScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
2.538 / 5
Higher measured value3.401 / 5
Vals AI
Source details
Data as of: 08/09/2026
CorpFin v2Shared Max Context
Test details
Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 128,000 tokens
Higher measured value65.734 %
61.888 %
Vals AI
Source details
Data as of: 08/06/2026
EMBDCF
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
Higher measured value46.388 %
44.435 %
Vals AI
Source details
Data as of: 08/06/2026
Finance Agent (v2)Comparables
Test details
Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
36.66 %
Higher measured value39.632 %
Vals AI
Source details
Data as of: 08/06/2026
Legal Research BenchBusiness & Commercial · All-pass
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 384,000 tokens
No direct winner23.256 %
No direct winner23.256 %
Vals AI
Source details
Data as of: 08/06/2026
LegalBenchIssue Tasks
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 256,000 tokens
Higher measured value91.48 %
84.587 %
Vals AI
Source details
Data as of: 08/09/2026
LiveCodeBench v6Medium
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGemini 3.1 Pro Preview: Reasoning: high, Temperature: 1, Output limit: 65,536 tokens
DeepSeek-V4-Pro: Reasoning: max, Output limit: 128,000 tokens
89.817 %
Higher measured value91.123 %
Vals AI
Source details
Data as of: 08/09/2026

Key to the marks:better valuebetter, but not conclusivetie

Our model tests

Community tests you can inspect

Compare how the models respond to the same prompt. Each test shows the first attempt, with no subsequent fixes to the generated code. These results do not contribute to an overall score.

Pelican on a bicycle

A community classic for free-form SVG drawing.

Gemini 3.1 Pro Preview

API access was unavailable for this run. The model was not evaluated.

First attempt
Run date
7 Sept 2026
API provider
openrouter
Requested API model
google/gemini-3.1-pro-preview
Pinned endpoint
Google AI Studio | google/gemini-3.1-pro-preview-20260219
Reasoning setting
high
Temperature
Provider default
Token limit including reasoning
8,192 tokens
Test protocol
community-visual-tests-v1
Technical checks
Not run

Not yet visually reviewed. A successful recording does not confirm correct physics or full compliance with the prompt.

DeepSeek-V4-Pro

API access was unavailable for this run. The model was not evaluated.

First attempt
Run date
7 Sept 2026
API provider
openrouter
Requested API model
deepseek/deepseek-v4-pro-0813
Pinned endpoint
DeepSeek | deepseek/deepseek-v4-pro-20260813
Reasoning setting
high
Temperature
Provider default
Token limit including reasoning
8,192 tokens
Test protocol
community-visual-tests-v1
Technical checks
Not run

Not yet visually reviewed. A successful recording does not confirm correct physics or full compliance with the prompt.

Prompt and test conditions

Generate an SVG of a pelican riding a bicycle

Token limit including reasoning: 8,192 tokens

Task origin (Simon Willison)

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.

The standard price table uses direct developer rates. The provider comparison labels OpenRouter endpoints, regions, and special tariffs separately.

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

Compare Gemini 3.1 Pro Preview and DeepSeek-V4-Pro with other leading models using the same data and benchmark logic.

View the complete LLM comparison