Skip to main content
BenchLM
Advertise

Model comparison

Claude Sonnet 4.6 vs Gemini 3.5 Flash

Data verified

Head-to-head evidence from 23 shared benchmark results across 7 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

65.07/100
Margin
0.3pts
← winning
64.75/100
2 category wins3 category wins

Public leaderboard positions: Claude Sonnet 4.6 #32 (Supported); Gemini 3.5 Flash #33 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Claude Sonnet 4.6 and Gemini 3.5 Flash share 23 comparable benchmark results. 5 of 8 categories are comparable. 10 results are unique to Claude Sonnet 4.6; 23 to Gemini 3.5 Flash.

Updated July 23, 2026
Shared results
23
Claude Sonnet 4.6 only
10
Gemini 3.5 Flash only
23
Comparable categories
5 / 8

Pick Claude Sonnet 4.6 if you want the stronger benchmark profile. Gemini 3.5 Flash only becomes the better choice if agentic is the priority or you want the cheaper token bill.

Confidence note. This is a partial-evidence comparison with 23 shared benchmark results across 7 evidence categories; 5 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Claude Sonnet 4.6 has the cleaner BenchAlign overall profile here, landing at 65.07 versus 64.75. It is a real lead, but still close enough that category-level strengths matter more than the headline number.

Claude Sonnet 4.6's sharpest advantage is in knowledge, where it averages 66 against 47.2. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 59.1% to 76.2%. Gemini 3.5 Flash does hit back in agentic, so the answer changes if that is the part of the workload you care about most.

Claude Sonnet 4.6 is also the more expensive model on tokens at $3.00 input / $15.00 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 Claude Sonnet 4.6 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. Gemini 3.5 Flash gives you the larger context window at 1M, compared with 200K for Claude Sonnet 4.6.

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 Claude Sonnet 4.6 and Gemini 3.5 Flash
CategoryClaude Sonnet 4.6ΔGemini 3.5 Flash
KnowledgeClaude Sonnet 4.666.0Margin← 18.8Gemini 3.5 Flash47.2
CodingClaude Sonnet 4.669.1Margin← 15.2Gemini 3.5 Flash53.9
AgenticClaude Sonnet 4.665.2Margin→ 12.0Gemini 3.5 Flash77.2
MathClaude Sonnet 4.626.4Margin→ 6.5Gemini 3.5 Flash32.9
MultimodalClaude Sonnet 4.677.4Margin→ 6.4Gemini 3.5 Flash83.8
ReasoningClaude Sonnet 4.6Not measuredMarginNo overlapGemini 3.5 Flash74.7
Inst. FollowingClaude Sonnet 4.6Not measuredMarginNo overlapGemini 3.5 Flash76.3

Decisive benchmark drivers

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

More
A · Claude Sonnet 4.6B · Gemini 3.5 Flash

  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 59.1%B 76.2%
    Winner: Gemini 3.5 FlashΔ 17.1
    Terminal-Bench 2.0: Claude Sonnet 4.6 scored 59.1%; Gemini 3.5 Flash scored 76.2%. Gemini 3.5 Flash wins this benchmark.

  2. HLE

    Knowledge
    Source ↗
    A 49%B 40.2%
    Winner: Claude Sonnet 4.6Δ 8.8
    HLE: Claude Sonnet 4.6 scored 49%; Gemini 3.5 Flash scored 40.2%. Claude Sonnet 4.6 wins this benchmark.

  3. CharXiv

    Multimodal
    Source ↗
    A 77.4%B 84.2%
    Winner: Gemini 3.5 FlashΔ 6.8
    CharXiv: Claude Sonnet 4.6 scored 77.4%; Gemini 3.5 Flash scored 84.2%. Gemini 3.5 Flash wins this benchmark.

  4. FrontierMath v2 (Tiers 1-3)

    Math
    Source ↗
    A 32.400%B 38.966%
    Winner: Gemini 3.5 FlashΔ 6.6
    FrontierMath v2 (Tiers 1-3): Claude Sonnet 4.6 scored 32.400%; Gemini 3.5 Flash scored 38.966%. Gemini 3.5 Flash wins this benchmark.

  5. OSWorld-Verified

    Agentic
    Source ↗
    A 72.1%B 78.4%
    Winner: Gemini 3.5 FlashΔ 6.3
    OSWorld-Verified: Claude Sonnet 4.6 scored 72.1%; Gemini 3.5 Flash scored 78.4%. Gemini 3.5 Flash wins this benchmark.

Operational comparison

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

MetricClaude Sonnet 4.6Gemini 3.5 FlashComparison
Input / output priceUSD per 1M tokensClaude Sonnet 4.6$3 input / $15 outputGemini 3.5 Flash$1.5 input / $9 outputGemini 3.5 Flash has the lower combined listed price.
Generation speedtokens per secondClaude Sonnet 4.644 tok/sGemini 3.5 Flash284.2 tok/sGemini 3.5 Flash has the higher measured throughput.
First-answer latencyseconds to first tokenClaude Sonnet 4.61.48 sGemini 3.5 Flash18.55 sClaude Sonnet 4.6 reaches the first token sooner.
Context windowmaximum listed tokensClaude Sonnet 4.6200KGemini 3.5 Flash1MGemini 3.5 Flash lists the larger context window.

Benchmark Deep Dive

AgenticGemini 3.5 Flash wins
BenchmarkClaude Sonnet 4.6Gemini 3.5 FlashResult
Terminal-Bench 2.0Source 59.1%76.2%Gemini 3.5 Flash leads
OSWorld-VerifiedSource 72.1%78.4%Gemini 3.5 Flash leads
Claw-EvalSource 67.8%—Not comparable
CyberGymSource 65.2%—Not comparable
τ²-bench resultsSource 79.5%95.3%Gemini 3.5 Flash leads
Gert LabsSource 62.92%61.85%Claude Sonnet 4.6 leads
OSWorld 2.0Source 8.3%—Not comparable
JobBenchSource 36.9%—Not comparable
MCP AtlasSource —83.6%Not comparable
ToolathlonSource —56.5%Not comparable
Finance Agent v2Source —57.9%Not comparable
GDPval-AASource —1349Not comparable
GDPval-AASource —42.4%Not comparable
AA Agentic IndexSource —37.5%Not comparable
APEX-Agents-AASource —47.1%Not comparable
ResearchClawBenchSource —18.0%Not comparable
AA AutomationBenchSource —42.6%Not comparable
AA EnterpriseOps-GymSource —50.1%Not comparable
terminalBenchHardSource —40.9%Not comparable
CodingClaude Sonnet 4.6 wins
BenchmarkClaude Sonnet 4.6Gemini 3.5 FlashResult
SWE-bench VerifiedSource 79.6%—Not comparable
SWE-RebenchSource 60.7%—Not comparable
React Native EvalsSource 80.6%—Not comparable
Vibe Code BenchSource 51.48%48.68%Claude Sonnet 4.6 leads
cursorBench31Source 48.8%49.8%Gemini 3.5 Flash leads
AA-SciCodeSource 46.9%53.1%Gemini 3.5 Flash leads
FrontierCode 1.1 MainSource 24.3%—Not comparable
Terminal-Bench 2.0Source —76.2%Not comparable
SWE-bench ProSource —55.1%Not comparable
SciCodeSource —53.1%Not comparable
cursorBench32Source —48.8%Not comparable
AA Coding IndexSource —70.1%Not comparable
Reasoning
BenchmarkClaude Sonnet 4.6Gemini 3.5 FlashResult
AA-LCRSource 57.7%69.3%Gemini 3.5 Flash leads
CritPtSource 0.9%13.1%Gemini 3.5 Flash leads
MRCRv2Source —77.3%Not comparable
MRCR 1MSource —26.6%Not comparable
ARC-AGI-2Source —72.1%Not comparable
KnowledgeClaude Sonnet 4.6 wins
BenchmarkClaude Sonnet 4.6Gemini 3.5 FlashResult
GPQASource 89.9%92.2%Gemini 3.5 Flash leads
SuperGPQASource 95%—Not comparable
MMLU-ProSource 79.2%—Not comparable
HLESource 49%40.2%Claude Sonnet 4.6 leads
Artificial Analysis Intelligence IndexSource 35.9%50.2%Gemini 3.5 Flash leads
AA-GPQA DiamondSource 79.9%92.2%Gemini 3.5 Flash leads
AA-HLESource 13.2%41.0%Gemini 3.5 Flash leads
AA-Omniscience IndexSource -2.9%22.7%Gemini 3.5 Flash leads
AA-Omniscience AccuracySource 38.0%51.9%Gemini 3.5 Flash leads
AA-Omniscience Hallucination RateSource 65.9%60.7%Gemini 3.5 Flash leads
GPQA-DSource —92.7%Not comparable
MathGemini 3.5 Flash wins
BenchmarkClaude Sonnet 4.6Gemini 3.5 FlashResult
FrontierMath v2 (Tiers 1-3)Source 32.400%38.966%Gemini 3.5 Flash leads
FrontierMath v2 (Tier 4)Source 8.300%14.583%Gemini 3.5 Flash leads
MultimodalGemini 3.5 Flash wins
BenchmarkClaude Sonnet 4.6Gemini 3.5 FlashResult
CharXivSource 77.4%84.2%Gemini 3.5 Flash leads
AA-MMMU-ProSource 70.6%84.3%Gemini 3.5 Flash leads
Design Arena WebsiteSource 13141285Claude Sonnet 4.6 leads
MMMU-ProSource —83.6%Not comparable
Blueprint-Bench 2Source —33.6%Not comparable
Inst. Following
BenchmarkClaude Sonnet 4.6Gemini 3.5 FlashResult
AA-IFBenchSource 41.2%76.3%Gemini 3.5 Flash leads
IFBenchSource —76.3%Not comparable
Frequently Asked Questions (6)

Which is better, Claude Sonnet 4.6 or Gemini 3.5 Flash?

Claude Sonnet 4.6 is ahead on BenchLM's BenchAlign leaderboard, 65.07 to 64.75. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 59.1% and 76.2%.

Which is better for knowledge tasks, Claude Sonnet 4.6 or Gemini 3.5 Flash?

Claude Sonnet 4.6 has the edge for knowledge tasks in this comparison, averaging 66 versus 47.2. Inside this category, AA-HLE is the benchmark that creates the most daylight between them.

Which is better for coding, Claude Sonnet 4.6 or Gemini 3.5 Flash?

Claude Sonnet 4.6 has the edge for coding in this comparison, averaging 69.1 versus 53.9. Inside this category, AA-SciCode is the benchmark that creates the most daylight between them.

Which is better for math, Claude Sonnet 4.6 or Gemini 3.5 Flash?

Gemini 3.5 Flash has the edge for math in this comparison, averaging 32.9 versus 26.4. Inside this category, FrontierMath v2 (Tiers 1-3) is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Claude Sonnet 4.6 or Gemini 3.5 Flash?

Gemini 3.5 Flash has the edge for agentic tasks in this comparison, averaging 77.2 versus 65.2. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.

Which is better for multimodal and grounded tasks, Claude Sonnet 4.6 or Gemini 3.5 Flash?

Gemini 3.5 Flash has the edge for multimodal and grounded tasks in this comparison, averaging 83.8 versus 77.4. Inside this category, Design Arena Website is the benchmark that creates the most daylight between them.

Related Comparisons

Explore More

Anthropic 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.