Skip to main content
BenchLM
Advertise

Model comparison

Gemini 3.1 Pro vs Gemini 3.5 Flash

Data verified

Head-to-head evidence from 30 shared benchmark results across 7 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

55.3/100
Margin
9.5pts
winning →
64.75/100
1 category wins2 category wins

Public leaderboard positions: Gemini 3.1 Pro #83 (Estimated); Gemini 3.5 Flash #33 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Gemini 3.1 Pro and Gemini 3.5 Flash share 30 comparable benchmark results. 3 of 8 categories are comparable. 17 results are unique to Gemini 3.1 Pro; 16 to Gemini 3.5 Flash.

Updated July 23, 2026
Shared results
30
Gemini 3.1 Pro only
17
Gemini 3.5 Flash only
16
Comparable categories
3 / 8

Pick Gemini 3.5 Flash if you want the stronger benchmark profile. Gemini 3.1 Pro only becomes the better choice if reasoning is the priority or you would rather avoid the extra latency and token burn of a reasoning model.

Confidence note. This is a partial-evidence comparison with 30 shared benchmark results across 7 evidence categories; 3 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Gemini 3.5 Flash is clearly ahead on the BenchAlign aggregate, 64.75 to 55.3. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Gemini 3.5 Flash's sharpest advantage is in multimodal & grounded, where it averages 83.8 against 82.6. The single biggest benchmark swing on the page is ARC-AGI-2, 77.1% to 72.1%. Gemini 3.1 Pro does hit back in reasoning, so the answer changes if that is the part of the workload you care about most.

Gemini 3.1 Pro is also the more expensive model on tokens at $2.00 input / $12.00 output per 1M tokens, versus $1.50 input / $9.00 output per 1M tokens for Gemini 3.5 Flash. Gemini 3.5 Flash is the reasoning model in the pair, while Gemini 3.1 Pro is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use.

Operator verdict

What I would choose

Gemini 3.5 Flash is the practical default here. It leads the public BenchAlign v5 lane 64.75 to 55.3, costs less at the listed API rates, and produced higher measured throughput. Choose Gemini 3.1 Pro only when its stronger results on specific reasoning tasks match your own workload.

Operator receipt. I checked 30 shared sourced results, the live BenchAlign v5 scores, and the operational rows on July 15, 2026. The receipt is concrete: $1.5 input / $9 output versus $2 input / $12 output, and 284.2 tok/s versus 109 tok/s.

Honest limit. Gemini 3.1 Pro's BenchAlign v5 score is an estimate with wider uncertainty, not a direct replay of the same production tasks. The benchmark and runtime rows narrow the choice; they do not replace an application-specific evaluation.

Google: Gemini 3.1 Pro Google Gemini API models BenchLM methodology

Reviewed by Glevd on July 15, 2026.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Category scores and score margins for Gemini 3.1 Pro and Gemini 3.5 Flash
CategoryGemini 3.1 ProΔGemini 3.5 Flash
ReasoningGemini 3.1 Pro77.1Margin← 2.4Gemini 3.5 Flash74.7
MultimodalGemini 3.1 Pro82.6Margin→ 1.2Gemini 3.5 Flash83.8
MathGemini 3.1 Pro31.8Margin→ 1.1Gemini 3.5 Flash32.9
AgenticGemini 3.1 ProNot measuredMarginNo overlapGemini 3.5 Flash77.2
CodingGemini 3.1 ProNot measuredMarginNo overlapGemini 3.5 Flash53.9
KnowledgeGemini 3.1 ProNot measuredMarginNo overlapGemini 3.5 Flash47.2
Inst. FollowingGemini 3.1 ProNot measuredMarginNo overlapGemini 3.5 Flash76.3

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

More
A · Gemini 3.1 ProB · Gemini 3.5 Flash

  1. ARC-AGI-2

    Reasoning
    Source ↗
    A 77.1%B 72.1%
    Winner: Gemini 3.1 ProΔ 5
    ARC-AGI-2: Gemini 3.1 Pro scored 77.1%; Gemini 3.5 Flash scored 72.1%. Gemini 3.1 Pro wins this benchmark.

  2. CharXiv

    Multimodal
    Source ↗
    A 80.2%B 84.2%
    Winner: Gemini 3.5 FlashΔ 4
    CharXiv: Gemini 3.1 Pro scored 80.2%; Gemini 3.5 Flash scored 84.2%. Gemini 3.5 Flash wins this benchmark.

  3. FrontierMath v2 (Tier 4)

    Math
    Source ↗
    A 16.700%B 14.583%
    Winner: Gemini 3.1 ProΔ 2.1
    FrontierMath v2 (Tier 4): Gemini 3.1 Pro scored 16.700%; Gemini 3.5 Flash scored 14.583%. Gemini 3.1 Pro wins this benchmark.

  4. FrontierMath v2 (Tiers 1-3)

    Math
    Source ↗
    A 36.900%B 38.966%
    Winner: Gemini 3.5 FlashΔ 2.1
    FrontierMath v2 (Tiers 1-3): Gemini 3.1 Pro scored 36.900%; Gemini 3.5 Flash scored 38.966%. Gemini 3.5 Flash wins this benchmark.

  5. MMMU-Pro

    Multimodal
    Source ↗
    A 83.9%B 83.6%
    Winner: Gemini 3.1 ProΔ 0.3
    MMMU-Pro: Gemini 3.1 Pro scored 83.9%; Gemini 3.5 Flash scored 83.6%. Gemini 3.1 Pro wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

MetricGemini 3.1 ProGemini 3.5 FlashComparison
Input / output priceUSD per 1M tokensGemini 3.1 Pro$2 input / $12 outputGemini 3.5 Flash$1.5 input / $9 outputGemini 3.5 Flash has the lower combined listed price.
Generation speedtokens per secondGemini 3.1 Pro109 tok/sGemini 3.5 Flash284.2 tok/sGemini 3.5 Flash has the higher measured throughput.
First-answer latencyseconds to first tokenGemini 3.1 Pro29.71 sGemini 3.5 Flash18.55 sGemini 3.5 Flash reaches the first token sooner.
Context windowmaximum listed tokensGemini 3.1 Pro1MGemini 3.5 Flash1MListed context windows are equal.

Benchmark Deep Dive

Agentic
BenchmarkGemini 3.1 ProGemini 3.5 FlashResult
Claw-EvalSource 57.8%—Not comparable
DeepSearchQASource 69.7%—Not comparable
τ²-bench resultsSource 95.6%95.3%Gemini 3.1 Pro leads
AA Agentic IndexSource 21.4%37.5%Gemini 3.5 Flash leads
APEX-Agents-AASource 32.0%47.1%Gemini 3.5 Flash leads
GDPval-AASource 23.3%42.4%Gemini 3.5 Flash leads
GDPval-AASource 9651349Gemini 3.5 Flash leads
Gert LabsSource 56.87%61.85%Gemini 3.5 Flash leads
ResearchClawBenchSource 13.3%18.0%Gemini 3.5 Flash leads
AA AutomationBenchSource 37.5%42.6%Gemini 3.5 Flash leads
AA EnterpriseOps-GymSource 42.2%50.1%Gemini 3.5 Flash leads
AA Harvey LABSource 58.9%—Not comparable
AA ITBenchSource 30.3%—Not comparable
AA Tau3 BankingSource 16.5%—Not comparable
terminalBenchHardSource 53.8%40.9%Gemini 3.1 Pro leads
aaTerminalBench21Source 73.8%—Not comparable
Terminal-Bench 2.0Source —76.2%Not comparable
MCP AtlasSource —83.6%Not comparable
ToolathlonSource —56.5%Not comparable
OSWorld-VerifiedSource —78.4%Not comparable
Finance Agent v2Source —57.9%Not comparable
Coding
BenchmarkGemini 3.1 ProGemini 3.5 FlashResult
LiveCodeBench ProSource 82.9%—Not comparable
React Native EvalsSource 78.9%—Not comparable
Vibe Code BenchSource 32.03%48.68%Gemini 3.5 Flash leads
AA Coding IndexSource 68.8%70.1%Gemini 3.5 Flash leads
AA-SciCodeSource 58.9%53.1%Gemini 3.1 Pro leads
Terminal-Bench 2.0Source —76.2%Not comparable
SWE-bench ProSource —55.1%Not comparable
SciCodeSource —53.1%Not comparable
cursorBench31Source —49.8%Not comparable
cursorBench32Source —48.8%Not comparable
ReasoningGemini 3.1 Pro wins
BenchmarkGemini 3.1 ProGemini 3.5 FlashResult
ARC-AGI-2Source 77.1%72.1%Gemini 3.1 Pro leads
AA-LCRSource 72.7%69.3%Gemini 3.1 Pro leads
CritPtSource 17.7%13.1%Gemini 3.1 Pro leads
MRCRv2Source —77.3%Not comparable
MRCR 1MSource —26.6%Not comparable
Knowledge
BenchmarkGemini 3.1 ProGemini 3.5 FlashResult
GPQA-DSource 94.3%92.7%Gemini 3.1 Pro leads
HLE w/o toolsSource 45.4%—Not comparable
HealthBench HardSource 20.6%—Not comparable
MedXpertQA (Text)Source 71.5%—Not comparable
Artificial Analysis Intelligence IndexSource 46.5%50.2%Gemini 3.5 Flash leads
AA-GPQA DiamondSource 94.1%92.2%Gemini 3.1 Pro leads
AA-HLESource 44.7%41.0%Gemini 3.1 Pro leads
AA-Omniscience IndexSource 32.9%22.7%Gemini 3.1 Pro leads
AA-Omniscience AccuracySource 55.3%51.9%Gemini 3.1 Pro leads
AA-Omniscience Hallucination RateSource 49.9%60.7%Gemini 3.1 Pro leads
GPQASource —92.2%Not comparable
HLESource —40.2%Not comparable
MathGemini 3.5 Flash wins
BenchmarkGemini 3.1 ProGemini 3.5 FlashResult
FrontierMath v2 (Tiers 1-3)Source 36.900%38.966%Gemini 3.5 Flash leads
FrontierMath v2 (Tier 4)Source 16.700%14.583%Gemini 3.1 Pro leads
Multilingual
BenchmarkGemini 3.1 ProGemini 3.5 FlashResult
AA Global-MMLU-LiteSource 93.2%—Not comparable
MultimodalGemini 3.5 Flash wins
BenchmarkGemini 3.1 ProGemini 3.5 FlashResult
MMMU-ProSource 83.9%83.6%Gemini 3.1 Pro leads
CharXivSource 80.2%84.2%Gemini 3.5 Flash leads
ERQASource 69.4%—Not comparable
SimpleVQASource 72.4%—Not comparable
ScreenSpot ProSource 84.4%—Not comparable
ZeroBenchSource 29.0%—Not comparable
MedXpertQA (MM)Source 81.3%—Not comparable
AA-MMMU-ProSource 82.4%84.3%Gemini 3.5 Flash leads
Design Arena WebsiteSource 12811285Gemini 3.5 Flash leads
Blueprint-Bench 2Source —33.6%Not comparable
Inst. Following
BenchmarkGemini 3.1 ProGemini 3.5 FlashResult
AA-IFBenchSource 77.1%76.3%Gemini 3.1 Pro leads
IFBenchSource —76.3%Not comparable
Frequently Asked Questions (4)

Which is better, Gemini 3.1 Pro or Gemini 3.5 Flash?

Gemini 3.5 Flash is ahead on BenchLM's BenchAlign leaderboard, 64.75 to 55.3. The biggest single separator in this matchup is ARC-AGI-2, where the scores are 77.1% and 72.1%.

Which is better for math, Gemini 3.1 Pro or Gemini 3.5 Flash?

Gemini 3.5 Flash has the edge for math in this comparison, averaging 32.9 versus 31.8. Inside this category, FrontierMath v2 (Tier 4) is the benchmark that creates the most daylight between them.

Which is better for reasoning, Gemini 3.1 Pro or Gemini 3.5 Flash?

Gemini 3.1 Pro has the edge for reasoning in this comparison, averaging 77.1 versus 74.7. Inside this category, ARC-AGI-2 is the benchmark that creates the most daylight between them.

Which is better for multimodal and grounded tasks, Gemini 3.1 Pro or Gemini 3.5 Flash?

Gemini 3.5 Flash has the edge for multimodal and grounded tasks in this comparison, averaging 83.8 versus 82.6. Inside this category, CharXiv is the benchmark that creates the most daylight between them.

Related Comparisons

Explore More

Google Compare Pricing Methodology Find Your Best LLM Overall Rankings
Last updated: July 23, 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.