Skip to main content
BenchLM
Advertise

Model comparison

Gemini 3.5 Flash vs Step 3.7 Flash

Data verified

Head-to-head evidence from 24 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

75.02/100
Margin
24.3pts
← winning
50.76/100
1 category wins1 category wins

Verified leaderboard positions: Gemini 3.5 Flash #7; Step 3.7 Flash unranked

BenchAlign evidence: Gemini 3.5 Flash supported; Step 3.7 Flash estimated. Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Gemini 3.5 Flash and Step 3.7 Flash share 24 comparable benchmark results. 2 of 8 categories are comparable. 27 results are unique to Gemini 3.5 Flash; 6 to Step 3.7 Flash.

Updated July 16, 2026
Shared results
24
Gemini 3.5 Flash only
27
Step 3.7 Flash only
6
Comparable categories
2 / 8

Pick Gemini 3.5 Flash if you want the stronger benchmark profile. Step 3.7 Flash only becomes the better choice if coding is the priority or you want the cheaper token bill.

Confidence note. This is a partial-evidence comparison with 24 shared benchmark results across 6 evidence categories; 2 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 provisional aggregate, 80 to 57. 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 agentic, where it averages 77.2 against 66.4. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 76.2% to 59.5%. Step 3.7 Flash does hit back in coding, so the answer changes if that is the part of the workload you care about most.

Gemini 3.5 Flash is also the more expensive model on tokens at $1.50 input / $9.00 output per 1M tokens, versus $0.20 input / $1.15 output per 1M tokens for Step 3.7 Flash. That is roughly 7.8x on output cost alone. Gemini 3.5 Flash gives you the larger context window at 1M, compared with 256K for Step 3.7 Flash.

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 3.5 Flash and Step 3.7 Flash
CategoryGemini 3.5 FlashΔStep 3.7 Flash
AgenticGemini 3.5 Flash77.2Margin← 10.8Step 3.7 Flash66.4
CodingGemini 3.5 Flash53.9Margin→ 2.4Step 3.7 Flash56.3
ReasoningGemini 3.5 Flash74.7MarginNo overlapStep 3.7 FlashNot measured
KnowledgeGemini 3.5 Flash47.2MarginNo overlapStep 3.7 FlashNot measured
MathGemini 3.5 Flash32.9MarginNo overlapStep 3.7 FlashNot measured
MultimodalGemini 3.5 Flash83.8MarginNo overlapStep 3.7 FlashNot measured
Inst. FollowingGemini 3.5 Flash76.3MarginNo overlapStep 3.7 FlashNot measured

Decisive benchmark drivers

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

More
A · Gemini 3.5 FlashB · Step 3.7 Flash

  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 76.2%B 59.5%
    Winner: Gemini 3.5 FlashΔ 16.7
    Terminal-Bench 2.0: Gemini 3.5 Flash scored 76.2%; Step 3.7 Flash scored 59.5%. Gemini 3.5 Flash wins this benchmark.

  2. SWE-bench Pro

    Coding
    Source ↗
    A 55.1%B 56.3%
    Winner: Step 3.7 FlashΔ 1.2
    SWE-bench Pro: Gemini 3.5 Flash scored 55.1%; Step 3.7 Flash scored 56.3%. Step 3.7 Flash wins this benchmark.

Operational comparison

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

MetricGemini 3.5 FlashStep 3.7 FlashComparison
Input / output priceUSD per 1M tokensGemini 3.5 Flash$1.5 input / $9 outputStep 3.7 Flash$0.2 input / $1.15 outputStep 3.7 Flash has the lower combined listed price.
Generation speedtokens per secondGemini 3.5 Flash284.2 tok/sStep 3.7 FlashNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenGemini 3.5 Flash18.55 sStep 3.7 FlashNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensGemini 3.5 Flash1MStep 3.7 Flash256KGemini 3.5 Flash lists the larger context window.

Benchmark Deep Dive

AgenticGemini 3.5 Flash wins
BenchmarkGemini 3.5 FlashStep 3.7 FlashResult
Terminal-Bench 2.0Source 76.2%59.5%Gemini 3.5 Flash leads
MCP AtlasSource 83.6%—Not comparable
ToolathlonSource 56.5%49.5%Gemini 3.5 Flash leads
OSWorld-VerifiedSource 78.4%—Not comparable
Finance Agent v2Source 57.9%—Not comparable
GDPval-AASource 13491017Gemini 3.5 Flash leads
τ²-bench resultsSource 95.3%98.5%Step 3.7 Flash leads
GDPval-AASource 42.4%25.9%Gemini 3.5 Flash leads
AA Agentic IndexSource 37.5%21.5%Gemini 3.5 Flash leads
APEX-Agents-AASource 47.1%14.8%Gemini 3.5 Flash leads
Gert LabsSource 62.80%51.57%Gemini 3.5 Flash leads
ResearchClawBenchSource 18.0%—Not comparable
AA BriefcaseSource 866—Not comparable
AA AutomationBenchSource 42.6%—Not comparable
AA EnterpriseOps-GymSource 50.1%—Not comparable
AA Harvey LABSource 1.7%—Not comparable
AA ITBenchSource 40.3%—Not comparable
AA Tau3 BankingSource 25.4%—Not comparable
BrowseCompSource —75.8%Not comparable
DeepSearchQASource —92.8%Not comparable
Claw-EvalSource —67.1%Not comparable
HLE w/ toolsSource —47.2%Not comparable
CodingStep 3.7 Flash wins
BenchmarkGemini 3.5 FlashStep 3.7 FlashResult
Terminal-Bench 2.0Source 76.2%59.5%Gemini 3.5 Flash leads
Terminal-Bench HardSource 40.9%35.6%Gemini 3.5 Flash leads
SWE-bench ProSource 55.1%56.3%Step 3.7 Flash leads
SciCodeSource 53.1%—Not comparable
Vibe Code BenchSource 48.68%—Not comparable
cursorBench31Source 49.8%—Not comparable
cursorBench32Source 48.8%—Not comparable
AA Coding IndexSource 70.1%39.6%Gemini 3.5 Flash leads
AA-SciCodeSource 53.1%40.0%Gemini 3.5 Flash leads
AA Terminal-Bench 2.1Source 78.7%—Not comparable
Reasoning
BenchmarkGemini 3.5 FlashStep 3.7 FlashResult
MRCRv2Source 77.3%—Not comparable
MRCR 1MSource 26.6%—Not comparable
ARC-AGI-2Source 72.1%—Not comparable
AA-LCRSource 69.3%63.7%Gemini 3.5 Flash leads
CritPtSource 13.1%2.3%Gemini 3.5 Flash leads
Knowledge
BenchmarkGemini 3.5 FlashStep 3.7 FlashResult
Artificial Analysis Intelligence IndexSource 50.2%30.3%Gemini 3.5 Flash leads
GPQASource 92.2%—Not comparable
GPQA-DSource 92.7%—Not comparable
HLESource 40.2%—Not comparable
AA-Omniscience AccuracySource 51.9%25.4%Gemini 3.5 Flash leads
AA-Omniscience Hallucination RateSource 60.7%84.4%Gemini 3.5 Flash leads
AA-GPQA DiamondSource 92.2%80.9%Gemini 3.5 Flash leads
AA-HLESource 41.0%19.9%Gemini 3.5 Flash leads
AA-Omniscience IndexSource 22.7%-37.5%Gemini 3.5 Flash leads
Math
BenchmarkGemini 3.5 FlashStep 3.7 FlashResult
FrontierMath v2 (Tiers 1-3)Source 38.966%—Not comparable
FrontierMath v2 (Tier 4)Source 14.583%—Not comparable
Multimodal
BenchmarkGemini 3.5 FlashStep 3.7 FlashResult
CharXivSource 84.2%—Not comparable
MMMU-ProSource 83.6%—Not comparable
Blueprint-Bench 2Source 33.6%—Not comparable
AA-MMMU-ProSource 84.3%75.3%Gemini 3.5 Flash leads
Design Arena WebsiteSource 12881218Gemini 3.5 Flash leads
SimpleVQASource —79.2%Not comparable
V*Source —95.3%Not comparable
Inst. Following
BenchmarkGemini 3.5 FlashStep 3.7 FlashResult
IFBenchSource 76.3%—Not comparable
AA-IFBenchSource 76.3%67.3%Gemini 3.5 Flash leads
Frequently Asked Questions (3)

Which is better, Gemini 3.5 Flash or Step 3.7 Flash?

Gemini 3.5 Flash is ahead on BenchLM's provisional leaderboard, 80 to 57. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 76.2% and 59.5%.

Which is better for coding, Gemini 3.5 Flash or Step 3.7 Flash?

Step 3.7 Flash has the edge for coding in this comparison, averaging 56.3 versus 53.9. Inside this category, AA Coding Index is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Gemini 3.5 Flash or Step 3.7 Flash?

Gemini 3.5 Flash has the edge for agentic tasks in this comparison, averaging 77.2 versus 66.4. Inside this category, GDPval-AA 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 16, 2026

The AI models change fast. We track them for you.

A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.