Skip to main content
BenchLM
Advertise

Model comparison

Qwen3.7 Max vs Qwen3.7 Plus

Data verified

Head-to-head evidence from 48 shared benchmark results across 8 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

72.84/100
Margin
5.6pts
← winning
67.22/100
4 category wins3 category wins

Public leaderboard positions: Qwen3.7 Max #10 (Supported); Qwen3.7 Plus #21 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Qwen3.7 Max and Qwen3.7 Plus share 48 comparable benchmark results. 7 of 8 categories are comparable. 10 results are unique to Qwen3.7 Max; 21 to Qwen3.7 Plus.

Updated July 23, 2026
Shared results
48
Qwen3.7 Max only
10
Qwen3.7 Plus only
21
Comparable categories
7 / 8

Pick Qwen3.7 Max if you want the stronger benchmark profile. Qwen3.7 Plus only becomes the better choice if agentic is the priority.

Confidence note. This is a partial-evidence comparison with 48 shared benchmark results across 8 evidence categories; 7 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Qwen3.7 Max is clearly ahead on the BenchAlign aggregate, 72.84 to 67.22. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Qwen3.7 Max's sharpest advantage is in mathematics, where it averages 97.1 against 92.9. The single biggest benchmark swing on the page is HLE, 41.4% to 34.7%. Qwen3.7 Plus does hit back in agentic, so the answer changes if that is the part of the workload you care about most.

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 Qwen3.7 Max and Qwen3.7 Plus
CategoryQwen3.7 MaxΔQwen3.7 Plus
MathQwen3.7 Max97.1Margin← 4.2Qwen3.7 Plus92.9
KnowledgeQwen3.7 Max64.2Margin← 4.1Qwen3.7 Plus60.1
CodingQwen3.7 Max77.9Margin← 2.3Qwen3.7 Plus75.6
AgenticQwen3.7 Max69.7Margin→ 2.0Qwen3.7 Plus71.7
MultilingualQwen3.7 Max87.0Margin← 1.6Qwen3.7 Plus85.4
ReasoningQwen3.7 Max90.4Margin→ 1.3Qwen3.7 Plus91.7
Inst. FollowingQwen3.7 Max84.4Margin→ 0.1Qwen3.7 Plus84.5
MultimodalQwen3.7 MaxNot measuredMarginNo overlapQwen3.7 Plus81.5

Decisive benchmark drivers

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

More
A · Qwen3.7 MaxB · Qwen3.7 Plus

  1. HLE

    Knowledge
    Source ↗
    A 41.4%B 34.7%
    Winner: Qwen3.7 MaxΔ 6.7
    HLE: Qwen3.7 Max scored 41.4%; Qwen3.7 Plus scored 34.7%. Qwen3.7 Max wins this benchmark.

  2. HMMT Feb 2026

    Math
    Source ↗
    A 97.1%B 92.9%
    Winner: Qwen3.7 MaxΔ 4.2
    HMMT Feb 2026: Qwen3.7 Max scored 97.1%; Qwen3.7 Plus scored 92.9%. Qwen3.7 Max wins this benchmark.

  3. SWE-bench Pro

    Coding
    Source ↗
    A 60.6%B 57.6%
    Winner: Qwen3.7 MaxΔ 3
    SWE-bench Pro: Qwen3.7 Max scored 60.6%; Qwen3.7 Plus scored 57.6%. Qwen3.7 Max wins this benchmark.

  4. SWE-bench Verified

    Coding
    Source ↗
    A 80.4%B 77.7%
    Winner: Qwen3.7 MaxΔ 2.7
    SWE-bench Verified: Qwen3.7 Max scored 80.4%; Qwen3.7 Plus scored 77.7%. Qwen3.7 Max wins this benchmark.

  5. SciCode

    Coding
    Source ↗
    A 53.5%B 51.3%
    Winner: Qwen3.7 MaxΔ 2.2
    SciCode: Qwen3.7 Max scored 53.5%; Qwen3.7 Plus scored 51.3%. Qwen3.7 Max wins this benchmark.

Operational comparison

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

MetricQwen3.7 MaxQwen3.7 PlusComparison
Input / output priceUSD per 1M tokensQwen3.7 MaxNot availableQwen3.7 PlusNot availableA complete price comparison is not available.
Generation speedtokens per secondQwen3.7 MaxNot availableQwen3.7 PlusNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenQwen3.7 MaxNot availableQwen3.7 PlusNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensQwen3.7 Max1MQwen3.7 Plus1MListed context windows are equal.

Benchmark Deep Dive

AgenticQwen3.7 Plus wins
BenchmarkQwen3.7 MaxQwen3.7 PlusResult
Terminal-Bench 2.0Source 69.7%70.3%Qwen3.7 Plus leads
QwenClawBenchSource 64.3%61.8%Qwen3.7 Max leads
QwenWebBenchSource 15681536Qwen3.7 Max leads
Claw-EvalSource 65.2%62.7%Qwen3.7 Max leads
BFCL v4Source 75.0%72.9%Qwen3.7 Max leads
MCP AtlasSource 76.4%73.2%Qwen3.7 Max leads
VITA-BenchSource 47.9%45.6%Qwen3.7 Max leads
HLE w/ toolsSource 53.5%—Not comparable
AA Agentic IndexSource 30.6%20.8%Qwen3.7 Max leads
τ²-bench resultsSource 94.7%93%Qwen3.7 Max leads
GDPval-AASource 38.7%21.8%Qwen3.7 Max leads
GDPval-AASource 1273936Qwen3.7 Max leads
Gert LabsSource 64.27%—Not comparable
ResearchClawBenchSource 18.7%—Not comparable
AA BriefcaseSource 908—Not comparable
AA AutomationBenchSource 25.6%—Not comparable
AA EnterpriseOps-GymSource 45.0%—Not comparable
AA ITBenchSource 42.5%—Not comparable
terminalBenchHardSource 50.8%—Not comparable
aaTerminalBench21Source 74.5%—Not comparable
AA Harvey LABSource 83.4%—Not comparable
DeepPlanningSource —62.3%Not comparable
OSWorld-VerifiedSource —73.3%Not comparable
AndroidWorldSource —81.0%Not comparable
APEX-Agents-AASource —22.4%Not comparable
OSWorld 2.0Source —2.8%Not comparable
CodingQwen3.7 Max wins
BenchmarkQwen3.7 MaxQwen3.7 PlusResult
SWE-bench VerifiedSource 80.4%77.7%Qwen3.7 Max leads
SWE-bench ProSource 60.6%57.6%Qwen3.7 Max leads
SWE MultilingualSource 78.3%75.8%Qwen3.7 Max leads
NL2RepoSource 47.2%41.1%Qwen3.7 Max leads
SciCodeSource 53.5%51.3%Qwen3.7 Max leads
LiveCodeBenchSource 91.6%89.6%Qwen3.7 Max leads
Terminal-Bench 2.0Source 69.7%70.3%Qwen3.7 Plus leads
AA Coding IndexSource 66.0%55.9%Qwen3.7 Max leads
AA-SciCodeSource 48.8%45.5%Qwen3.7 Max leads
ReasoningQwen3.7 Plus wins
BenchmarkQwen3.7 MaxQwen3.7 PlusResult
MRCRv2Source 90.4%91.7%Qwen3.7 Plus leads
CritPtSource 13.4%9.1%Qwen3.7 Max leads
AA-LCRSource 69.0%65.0%Qwen3.7 Max leads
KnowledgeQwen3.7 Max wins
BenchmarkQwen3.7 MaxQwen3.7 PlusResult
GPQASource 92.4%90.3%Qwen3.7 Max leads
GPQA-DSource 92.4%90.3%Qwen3.7 Max leads
HLESource 41.4%34.7%Qwen3.7 Max leads
MMLU-ProSource 89.6%88.5%Qwen3.7 Max leads
MMLU-ReduxSource 95%94.5%Qwen3.7 Max leads
SuperGPQASource 73.6%71.4%Qwen3.7 Max leads
MMMLUSource 90.3%89.0%Qwen3.7 Max leads
Artificial Analysis Intelligence IndexSource 46.0%39.0%Qwen3.7 Max leads
AA-GPQA DiamondSource 92.3%90.0%Qwen3.7 Max leads
AA-HLESource 38.1%33.4%Qwen3.7 Max leads
AA-Omniscience IndexSource 14.1%2.4%Qwen3.7 Max leads
AA-Omniscience AccuracySource 30.1%22.2%Qwen3.7 Max leads
AA-Omniscience Hallucination RateSource 22.9%25.5%Qwen3.7 Max leads
MathQwen3.7 Max wins
BenchmarkQwen3.7 MaxQwen3.7 PlusResult
HMMT Feb 2026Source 97.1%92.9%Qwen3.7 Max leads
IMOAnswerBenchSource 90.0%86.0%Qwen3.7 Max leads
ApexSource 44.5%22.7%Qwen3.7 Max leads
MultilingualQwen3.7 Max wins
BenchmarkQwen3.7 MaxQwen3.7 PlusResult
MMLU-ProXSource 87%85.4%Qwen3.7 Max leads
NOVA-63Source 59.0%58.8%Qwen3.7 Max leads
INCLUDESource 86.2%83.0%Qwen3.7 Max leads
MAXIFESource 89.2%88.8%Qwen3.7 Max leads
PolyMathSource 86.5%84.0%Qwen3.7 Max leads
Multimodal
BenchmarkQwen3.7 MaxQwen3.7 PlusResult
Design Arena WebsiteSource 12931288Qwen3.7 Max leads
MMMU-ProSource —79%Not comparable
MathVisionSource —90.3%Not comparable
CharXivSource —85.9%Not comparable
ERQASource —69.8%Not comparable
MedXpertQA (MM)Source —71.0%Not comparable
ScreenSpot ProSource —79.0%Not comparable
SimpleVQASource —81.7%Not comparable
MMSearch-PlusSource —41.4%Not comparable
RealWorldQASource —86.9%Not comparable
OmniDocBench 1.5Source —91.4%Not comparable
OCRBench V2Source —70.7%Not comparable
ODINW13Source —51.1%Not comparable
Video-MME (with subtitle)Source —88.0%Not comparable
VideoMMMUSource —85.4%Not comparable
MLVU (M-Avg)Source —87.4%Not comparable
AA-MMMU-ProSource —80.5%Not comparable
Inst. FollowingQwen3.7 Plus wins
BenchmarkQwen3.7 MaxQwen3.7 PlusResult
IFEvalSource 94.3%94.6%Qwen3.7 Plus leads
IFBenchSource 79.1%79.1%Tie
AA-IFBenchSource 80.5%78.0%Qwen3.7 Max leads
Frequently Asked Questions (8)

Which is better, Qwen3.7 Max or Qwen3.7 Plus?

Qwen3.7 Max is ahead on BenchLM's BenchAlign leaderboard, 72.84 to 67.22. The biggest single separator in this matchup is HLE, where the scores are 41.4% and 34.7%.

Which is better for knowledge tasks, Qwen3.7 Max or Qwen3.7 Plus?

Qwen3.7 Max has the edge for knowledge tasks in this comparison, averaging 64.2 versus 60.1. Inside this category, AA-Omniscience Index is the benchmark that creates the most daylight between them.

Which is better for coding, Qwen3.7 Max or Qwen3.7 Plus?

Qwen3.7 Max has the edge for coding in this comparison, averaging 77.9 versus 75.6. Inside this category, AA Coding Index is the benchmark that creates the most daylight between them.

Which is better for math, Qwen3.7 Max or Qwen3.7 Plus?

Qwen3.7 Max has the edge for math in this comparison, averaging 97.1 versus 92.9. Inside this category, Apex is the benchmark that creates the most daylight between them.

Which is better for reasoning, Qwen3.7 Max or Qwen3.7 Plus?

Qwen3.7 Plus has the edge for reasoning in this comparison, averaging 91.7 versus 90.4. Inside this category, CritPt is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Qwen3.7 Max or Qwen3.7 Plus?

Qwen3.7 Plus has the edge for agentic tasks in this comparison, averaging 71.7 versus 69.7. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.

Which is better for instruction following, Qwen3.7 Max or Qwen3.7 Plus?

Qwen3.7 Plus has the edge for instruction following in this comparison, averaging 84.5 versus 84.4. Inside this category, AA-IFBench is the benchmark that creates the most daylight between them.

Which is better for multilingual tasks, Qwen3.7 Max or Qwen3.7 Plus?

Qwen3.7 Max has the edge for multilingual tasks in this comparison, averaging 87 versus 85.4. Inside this category, INCLUDE is the benchmark that creates the most daylight between them.

Related Comparisons

Explore More

Alibaba 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.