Skip to main content
BenchLM
Advertise

Model comparison

Gemini 2.5 Pro vs Qwen3.5 397B

Data verified

Head-to-head evidence from 20 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

57.25/100
Margin
0.2pts
← winning
57.01/100
0 category wins3 category wins

Public leaderboard positions: Gemini 2.5 Pro #70 (Supported); Qwen3.5 397B #71 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Gemini 2.5 Pro and Qwen3.5 397B share 20 comparable benchmark results. 3 of 8 categories are comparable. 4 results are unique to Gemini 2.5 Pro; 35 to Qwen3.5 397B.

Updated July 22, 2026
Shared results
20
Gemini 2.5 Pro only
4
Qwen3.5 397B only
35
Comparable categories
3 / 8

Pick Gemini 2.5 Pro if you want the stronger benchmark profile. Qwen3.5 397B only becomes the better choice if mathematics is the priority or you want the cheaper token bill.

Confidence note. This is a partial-evidence comparison with 20 shared benchmark results across 6 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 2.5 Pro has the cleaner BenchAlign overall profile here, landing at 57.25 versus 57.01. It is a real lead, but still close enough that category-level strengths matter more than the headline number.

Gemini 2.5 Pro is also the more expensive model on tokens at $1.25 input / $10.00 output per 1M tokens, versus $0.60 input / $3.60 output per 1M tokens for Qwen3.5 397B. That is roughly 2.8x on output cost alone. Gemini 2.5 Pro gives you the larger context window at 1M, compared with 128K for Qwen3.5 397B.

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 2.5 Pro and Qwen3.5 397B
CategoryGemini 2.5 ProΔQwen3.5 397B
MathGemini 2.5 Pro11.6Margin→ 79.0Qwen3.5 397B90.6
KnowledgeGemini 2.5 Pro27.4Margin→ 29.2Qwen3.5 397B56.6
CodingGemini 2.5 Pro63.8Margin→ 2.7Qwen3.5 397B66.5
AgenticGemini 2.5 ProNot measuredMarginNo overlapQwen3.5 397B56.5
ReasoningGemini 2.5 ProNot measuredMarginNo overlapQwen3.5 397B63.2
MultilingualGemini 2.5 ProNot measuredMarginNo overlapQwen3.5 397B84.7
MultimodalGemini 2.5 ProNot measuredMarginNo overlapQwen3.5 397B79.6
Inst. FollowingGemini 2.5 ProNot measuredMarginNo overlapQwen3.5 397B92.6

Decisive benchmark drivers

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

More
A · Gemini 2.5 ProB · Qwen3.5 397B

  1. SWE-bench Verified

    Coding
    Source ↗
    A 63.8%B 76.2%
    Winner: Qwen3.5 397BΔ 12.4
    SWE-bench Verified: Gemini 2.5 Pro scored 63.8%; Qwen3.5 397B scored 76.2%. Qwen3.5 397B wins this benchmark.

  2. HLE

    Knowledge
    Source ↗
    A 18.8%B 28.7%
    Winner: Qwen3.5 397BΔ 9.9
    HLE: Gemini 2.5 Pro scored 18.8%; Qwen3.5 397B scored 28.7%. Qwen3.5 397B wins this benchmark.

  3. GPQA

    Knowledge
    Source ↗
    A 83%B 88.4%
    Winner: Qwen3.5 397BΔ 5.4
    GPQA: Gemini 2.5 Pro scored 83%; Qwen3.5 397B scored 88.4%. Qwen3.5 397B wins this benchmark.

Operational comparison

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

MetricGemini 2.5 ProQwen3.5 397BComparison
Input / output priceUSD per 1M tokensGemini 2.5 Pro$1.25 input / $10 outputQwen3.5 397B$0.6 input / $3.6 outputQwen3.5 397B has the lower combined listed price.
Generation speedtokens per secondGemini 2.5 Pro117 tok/sQwen3.5 397B96 tok/sGemini 2.5 Pro has the higher measured throughput.
First-answer latencyseconds to first tokenGemini 2.5 Pro21.19 sQwen3.5 397B2.44 sQwen3.5 397B reaches the first token sooner.
Context windowmaximum listed tokensGemini 2.5 Pro1MQwen3.5 397B128KGemini 2.5 Pro lists the larger context window.

Benchmark Deep Dive

Agentic
BenchmarkGemini 2.5 ProQwen3.5 397BResult
AA Agentic IndexSource 7.1%19.9%Qwen3.5 397B leads
τ²-bench resultsSource 54.1%95.6%Qwen3.5 397B leads
Gert LabsSource 42.01%46.76%Qwen3.5 397B leads
GDPval-AASource 8.3%23.1%Qwen3.5 397B leads
GDPval-AASource 665962Qwen3.5 397B leads
Terminal-Bench 2.0Source —52.5%Not comparable
BrowseCompSource —62%Not comparable
Claw-EvalSource —56.8%Not comparable
QwenClawBenchSource —51.8%Not comparable
τ³-bench resultsSource —68.4%Not comparable
VITA-BenchSource —43.7%Not comparable
DeepPlanningSource —37.6%Not comparable
ToolathlonSource —36.3%Not comparable
MCP AtlasSource —46.1%Not comparable
MCP-TasksSource —74.2%Not comparable
WideResearchSource —74.0%Not comparable
ResearchClawBenchSource —14.2%Not comparable
APEX-Agents-AASource —15.3%Not comparable
CodingQwen3.5 397B wins
BenchmarkGemini 2.5 ProQwen3.5 397BResult
SWE-bench VerifiedSource 63.8%76.2%Qwen3.5 397B leads
Vibe Code BenchSource 0.40%—Not comparable
AA Coding IndexSource 33.3%48.2%Qwen3.5 397B leads
AA-SciCodeSource 42.8%42.0%Gemini 2.5 Pro leads
LiveCodeBench v6Source —83.6%Not comparable
SWE-bench ProSource —50.9%Not comparable
Reasoning
BenchmarkGemini 2.5 ProQwen3.5 397BResult
AA-LCRSource 66.0%65.7%Gemini 2.5 Pro leads
CritPtSource 2.6%1.7%Gemini 2.5 Pro leads
LongBench v2Source —63.2%Not comparable
AI-NeedleSource —68.7%Not comparable
KnowledgeQwen3.5 397B wins
BenchmarkGemini 2.5 ProQwen3.5 397BResult
GPQASource 83%88.4%Qwen3.5 397B leads
HLESource 18.8%28.7%Qwen3.5 397B leads
Artificial Analysis Intelligence IndexSource 25.8%33.7%Qwen3.5 397B leads
AA-GPQA DiamondSource 84.4%89.3%Qwen3.5 397B leads
AA-HLESource 21.1%27.3%Qwen3.5 397B leads
AA-Omniscience IndexSource -14.3%-29.8%Gemini 2.5 Pro leads
AA-Omniscience AccuracySource 39.0%31.4%Gemini 2.5 Pro leads
AA-Omniscience Hallucination RateSource 87.4%89.1%Gemini 2.5 Pro leads
SuperGPQASource —70.4%Not comparable
MMLU-ProSource —87.8%Not comparable
MMLU-ReduxSource —94.9%Not comparable
C-EvalSource —93%Not comparable
MathQwen3.5 397B wins
BenchmarkGemini 2.5 ProQwen3.5 397BResult
FrontierMath v2 (Tiers 1-3)Source 14.138%—Not comparable
FrontierMath v2 (Tier 4)Source 4.167%—Not comparable
AIME26Source —93.3%Not comparable
HMMT Feb 2025Source —94.8%Not comparable
HMMT Nov 2025Source —92.7%Not comparable
HMMT Feb 2026Source —87.9%Not comparable
MMAnswerBenchSource —80.9%Not comparable
Multilingual
BenchmarkGemini 2.5 ProQwen3.5 397BResult
MMLU-ProXSource —84.7%Not comparable
NOVA-63Source —59.1%Not comparable
Multimodal
BenchmarkGemini 2.5 ProQwen3.5 397BResult
AA-MMMU-ProSource 74.9%77.3%Qwen3.5 397B leads
Design Arena WebsiteSource 1197—Not comparable
MMMU-ProSource —79%Not comparable
MathVisionSource —88.6%Not comparable
CharXivSource —80.8%Not comparable
VideoMMMUSource —84.7%Not comparable
ScreenSpot ProSource —65.6%Not comparable
V*Source —95.8%Not comparable
Inst. Following
BenchmarkGemini 2.5 ProQwen3.5 397BResult
AA-IFBenchSource 48.7%78.8%Qwen3.5 397B leads
IFEvalSource —92.6%Not comparable
Frequently Asked Questions (4)

Which is better, Gemini 2.5 Pro or Qwen3.5 397B?

Gemini 2.5 Pro is ahead on BenchLM's BenchAlign leaderboard, 57.25 to 57.01. The biggest single separator in this matchup is SWE-bench Verified, where the scores are 63.8% and 76.2%.

Which is better for knowledge tasks, Gemini 2.5 Pro or Qwen3.5 397B?

Qwen3.5 397B has the edge for knowledge tasks in this comparison, averaging 56.6 versus 27.4. Inside this category, AA-Omniscience Index is the benchmark that creates the most daylight between them.

Which is better for coding, Gemini 2.5 Pro or Qwen3.5 397B?

Qwen3.5 397B has the edge for coding in this comparison, averaging 66.5 versus 63.8. Inside this category, AA Coding Index is the benchmark that creates the most daylight between them.

Which is better for math, Gemini 2.5 Pro or Qwen3.5 397B?

Qwen3.5 397B has the edge for math in this comparison, averaging 90.6 versus 11.6. Gemini 2.5 Pro stays close enough that the answer can still flip depending on your workload.

Related Comparisons

Explore More

Google Compare Pricing Methodology Find Your Best LLM Overall Rankings
Last updated: July 22, 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.