Skip to main content

Model profile

Qwen3.7 Plus

AlibabaCurrentReleased Jun 3, 2026
Data verified
Overall Score
67.22Public #21 of 200Verified #19 of 99
Arena Elo
1461
Eligible category ranks
6of 8
Price (1M tokens)
Not listedAPI pricing
Speed
Not listed
Context
1M

Evidence coverage

69 of 323 tracked benchmarks are published. 52 are verified and 17 provisional. 8 of 8 categories are measured.

Updated July 23, 2026Methodology
Published / tracked
69 / 323
Verified
52
Provisional
17
Categories with evidence
8 / 8

Evidence by category

  • Agentic16 benchmarks
    Mixed evidence
  • Coding9 benchmarks
    Mixed evidence
  • Reasoning3 benchmarks
    Mixed evidence
  • Knowledge13 benchmarks
    Mixed evidence
  • Math3 benchmarks
    Verified
  • Multilingual5 benchmarks
    Verified
  • Multimodal17 benchmarks
    Mixed evidence
  • Inst. Following3 benchmarks
    Mixed evidence
ProprietaryReasoning
Confidence:
Very high
base

Qwen3.7 Plus ranks #21 out of 200 models on the public leaderboard with an overall score of 67.22/100. It also ranks #19 out of 99 on the verified leaderboard. This places it in the mid-tier of AI models, with strengths in specific benchmark categories.

Qwen3.7 Plus is a proprietary model with a 1M token context window. It uses explicit chain-of-thought reasoning, which typically improves performance on math and complex reasoning tasks at the cost of higher latency and token usage.

BenchLM links it directly to Qwen3.6 Plus as the earlier related model in that lineage. This profile currently has 69 of 323 tracked benchmarks. BenchLM only exposes non-generated benchmark rows publicly, so missing categories stay blank until a sourced evaluation is available.

Its strongest category is Multilingual (#3), while its weakest is Agentic (#81). This performance profile makes it a well-rounded choice across a range of tasks.

Peer position

Exact provisional scores and ranks for the closest listed peers. A score can appear before a model clears the evidence threshold for a rank, so equal scores can have different rank states.

Range 66.7967.73

  1. Gemini 3 Pro
    Google
    #1967.73
    Gemini 3 Pro is #19 with a score of 67.73.
    Compare
  2. Inkling
    Thinking Machines Lab
    #2067.54
    Inkling is #20 with a score of 67.54.
    Compare
  3. Qwen3.7 PlusCurrent model
    Alibaba
    #2167.22
    Qwen3.7 Plus is #21 with a score of 67.22.
  4. GPT-5.6 Luna
    OpenAI
    #2267.17
    GPT-5.6 Luna is #22 with a score of 67.17.
    Compare
  5. GPT-5.2 Pro
    OpenAI
    #2367.01
    GPT-5.2 Pro is #23 with a score of 67.01.
    Compare
  6. GLM-5-Turbo
    Z.AI
    #2466.89
    GLM-5-Turbo is #24 with a score of 66.89.
    Compare
  7. GPT-5.4 nano
    OpenAI
    #2566.79
    GPT-5.4 nano is #25 with a score of 66.79.
    Compare

Category percentile

More

Relative position among models eligible for each sourced category. A higher percentile means a stronger position within that category's ranked cohort; 100 is highest.

  1. Multilingual83%
    Eligible cohort rank #3 of 13Category score 78.9
  2. Inst. Following87%
    Eligible cohort rank #5 of 31Category score 91.4
  3. Multimodal68%
    Eligible cohort rank #10 of 29Category score 71.5
  4. Coding79%
    Eligible cohort rank #27 of 122Category score 58.8
  5. Knowledge33%
    Eligible cohort rank #35 of 52Category score 65.1
  6. Agentic32%
    Eligible cohort rank #81 of 119Category score 43.9

Category evidence

Scores and ranks appear only where this model has published benchmark evidence. Categories without displayable source records remain not measured.

Category scores, ranks, weighting, benchmark coverage, and evidence status
CategoryScore
AgenticRank #81 of 119Percentile 32ndWeight 22%16 benchmarksMixed sources43.9
CodingRank #27 of 122Percentile 79thWeight 20%9 benchmarksMixed sources58.8
ReasoningRank Not rankedWeight 17%3 benchmarksMixed sources87.1
KnowledgeRank #35 of 52Percentile 33rdWeight 12%13 benchmarksMixed sources65.1
MathRank Not rankedWeight 5%3 benchmarksVerified79.6
MultilingualRank #3 of 13Percentile 83rdWeight 7%5 benchmarksVerified78.9
MultimodalRank #10 of 29Percentile 68thWeight 12%17 benchmarksMixed sources71.5
Inst. FollowingRank #5 of 31Percentile 87thWeight 5%3 benchmarksMixed sources91.4

Chatbot Arena performance

Scroll horizontally to inspect confidence intervals and vote counts.

Chatbot Arena Elo, confidence interval, and vote count by evaluation view
ViewEloConfidence intervalVotes
Text Overall1461±5.621,513
Coding1510±8.46,298
Math1474±18.7996
Instruction Following1450±7.87,337
Creative Writing1439±10.43,636
Multi-turn1473±10.43,612
Hard Prompts1479±6.514,310
Hard Prompts (English)1482±8.26,601
Longer Query1469±7.59,441

Benchmark Details

Rows below have a displayable published verification record. Each source link and provenance note remains in the page HTML while its category is closed. Source-unverified manual rows and generated rows stay hidden.

Agentic16 benchmarks
Terminal-Bench 2.0Provider exact
70.3%Weighted 38%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Alibaba Cloud reports Terminal-Bench 2.0-Terminus at 70.3.
OSWorld-VerifiedProvider exact
73.3%Weighted 34%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
QwenClawBenchProvider exact
61.8%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
QwenWebBenchProvider exact
1536Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Alibaba Cloud reports QwenWebDev Elo at 1536. BenchLM stores it on the existing Qwen web agent benchmark key.
Claw-EvalProvider exact
62.7%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
BFCL v4Provider exact

Berkeley Function Calling Leaderboard v4

72.9%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
MCP AtlasProvider exact
73.2%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
VITA-BenchProvider exact
45.6%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
DeepPlanningProvider exact
62.3%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
AndroidWorldProvider exact
81.0%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
AA Agentic IndexReported

Artificial Analysis Agentic Index

20.8%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
APEX-Agents-AAReported
22.4%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
τ²-bench resultsReported

τ²-Bench Tool-Agent-User Evaluation

93%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
GDPval-AAReported

GDPval-AA normalized

21.8%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
GDPval-AAReported
936Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
OSWorld 2.0Benchmark exact
2.8%Display only
Source: OSWorld 2.0 paperProvenance: OSWorld 2.0 reports Qwen 3.7-Plus single action on its 500-step main table. BenchLM stores the binary completion score and notes the corresponding partial score was 21.5%.
Coding9 benchmarks
LiveCodeBenchProvider exact

LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code

89.6%Weighted 38%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
SWE-bench VerifiedProvider exact

Software Engineering Benchmark Verified

77.7%Weighted 16%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
SciCodeProvider exact

Scientific Code Benchmark

51.3%Weighted 16%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
SWE-bench ProProvider exact
57.6%Weighted 10%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
Terminal-Bench 2.0Provider exact
70.3%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Alibaba Cloud reports Terminal-Bench 2.0-Terminus at 70.3. BenchLM stores it in both coding and agentic views.
SWE MultilingualProvider exact
75.8%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
NL2RepoProvider exact
41.1%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
AA Coding IndexReported

Artificial Analysis Coding Index

55.9%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-SciCodeReported

Artificial Analysis SciCode

45.5%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Reasoning3 benchmarks
MRCRv2Provider exact
91.7%Weighted 31%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Alibaba Cloud reports MRCR-v2 128k at 91.7. BenchLM stores it on the existing MRCR-v2 key.
CritPtProvider exact

Critical Physics Tasks

9.1%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
AA-LCRReported

Artificial Analysis Long Context Reasoning

65.0%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Knowledge13 benchmarks
HLEProvider exact

Humanity's Last Exam

34.7%Weighted 45%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
MMLU-ProProvider exact

Massive Multitask Language Understanding Professional

88.5%Weighted 30%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
GPQAProvider exact

Graduate-Level Google-Proof Q&A

90.3%Weighted 7%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Alibaba Cloud reports GPQA Diamond at 90.3. BenchLM stores that exact value on the weighted GPQA lane and the display GPQA-Diamond lane.
SuperGPQAProvider exact

SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines

71.4%Weighted 7%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
GPQA-DProvider exact

GPQA Diamond

90.3%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Alibaba Cloud reports GPQA Diamond at 90.3.
MMLU-ReduxProvider exact
94.5%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
MMMLUProvider exact
89.0%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
Artificial Analysis Intelligence IndexReported
39.0%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-GPQA DiamondReported

Artificial Analysis GPQA Diamond

90.0%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-HLEReported

Artificial Analysis Humanity's Last Exam

33.4%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-Omniscience IndexReported

Artificial Analysis Omniscience Index

2.4%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-Omniscience AccuracyReported

Artificial Analysis Omniscience Accuracy

22.2%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-Omniscience Hallucination RateReported

Artificial Analysis Omniscience Hallucination Rate

25.5%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Math3 benchmarks
HMMT Feb 2026Provider exact

Harvard-MIT Mathematics Tournament February 2026

92.9%Weighted 25%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
IMOAnswerBenchProvider exact
86.0%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
ApexProvider exact
22.7%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
Multilingual5 benchmarks
MMLU-ProXProvider exact
85.4%Weighted 100%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
NOVA-63Provider exact
58.8%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
INCLUDEProvider exact
83.0%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
MAXIFEProvider exact
88.8%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
PolyMathProvider exact
84.0%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
Multimodal17 benchmarks
MMMU-ProProvider exact

Massive Multi-discipline Multimodal Understanding Pro

79%Weighted 45%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
CharXivProvider exact

CharXiv Reasoning

85.9%Weighted 25%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Alibaba Cloud reports CharXiv(RQ) as 85.9 with CI / 84.4 without CI. BenchLM stores the with-CI value on the existing CharXiv key.
MathVisionProvider exact
90.3%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
ERQAProvider exact
69.8%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
MedXpertQA (MM)Provider exact

MedXpertQA Multimodal

71.0%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
ScreenSpot ProProvider exact
79.0%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
SimpleVQAProvider exact
81.7%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
MMSearch-PlusProvider exact
41.4%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
RealWorldQAProvider exact
86.9%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
OmniDocBench 1.5Provider exact
91.4%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
OCRBench V2Provider exact
70.7%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Alibaba Cloud reports OCR-Bench-V2 (EN) at 70.7. BenchLM stores the English row on the existing OCR-Bench-V2 key.
ODINW13Provider exact
51.1%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
Video-MME (with subtitle)Provider exact

Video-MME with subtitle

88.0%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
VideoMMMUProvider exact
85.4%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
MLVU (M-Avg)Provider exact

MLVU mean average

87.4%Display only
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
AA-MMMU-ProReported

Artificial Analysis MMMU-Pro

80.5%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Design Arena WebsiteReported

Design Arena Website Elo

1288Display only
Source: OpenRouter model benchmarksProvenance: Display-only Design Arena Website Elo synced from OpenRouter model benchmark metadata. It is excluded from BenchLM weighted scoring.
Inst. Following3 benchmarks
IFBenchProvider exact

Instruction Following Benchmark

79.1%Weighted 65%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
IFEvalProvider exact

Instruction-Following Eval

94.6%Weighted 35%
Source: Alibaba Cloud: Qwen3.7-Plus launch benchmarksProvenance: Provider exact
AA-IFBenchReported

Artificial Analysis IFBench

78.0%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.

Qwen3.7 Plus Family

Base entry

Related Earlier Model

Qwen3.6 Plus

Frequently Asked Questions

How does Qwen3.7 Plus perform overall in AI benchmarks?

Qwen3.7 Plus currently ranks #21 out of 200 models on BenchLM's provisional leaderboard with an overall score of 67.22. It also ranks #19 out of 99 on the verified leaderboard. It is created by Alibaba. Its published context window is 1M.

Is Qwen3.7 Plus good for knowledge and understanding?

Qwen3.7 Plus ranks #35 out of 52 models in knowledge and understanding benchmarks with an average score of 65.1. There are stronger options in this category.

Is Qwen3.7 Plus good for coding and programming?

Qwen3.7 Plus ranks #27 out of 122 models in coding and programming benchmarks with an average score of 58.8. There are stronger options in this category.

Is Qwen3.7 Plus good for mathematics?

Qwen3.7 Plus has visible benchmark coverage in mathematics, but BenchLM does not currently assign it a global category rank there.

Is Qwen3.7 Plus good for reasoning and logic?

Qwen3.7 Plus has visible benchmark coverage in reasoning and logic, but BenchLM does not currently assign it a global category rank there.

Is Qwen3.7 Plus good for agentic tool use and computer tasks?

Qwen3.7 Plus ranks #81 out of 119 models in agentic tool use and computer tasks benchmarks with an average score of 43.9. There are stronger options in this category.

Is Qwen3.7 Plus good for multimodal and grounded tasks?

Qwen3.7 Plus ranks #10 out of 29 models in multimodal and grounded tasks benchmarks with an average score of 71.5. It is among the top performers in this category.

Is Qwen3.7 Plus good for instruction following?

Qwen3.7 Plus ranks #5 out of 31 models in instruction following benchmarks with an average score of 91.4. It is among the top performers in this category.

Is Qwen3.7 Plus good for multilingual tasks?

Qwen3.7 Plus ranks #3 out of 13 models in multilingual tasks benchmarks with an average score of 78.9. It is among the top performers in this category.

Does Qwen3.7 Plus have full benchmark coverage on BenchLM?

Not yet. Qwen3.7 Plus currently has 69 published benchmark scores out of the 323 benchmarks BenchLM tracks. BenchLM only exposes non-generated public benchmark rows, so missing categories stay blank until a sourced evaluation is available.

What is the context window size of Qwen3.7 Plus?

Qwen3.7 Plus has a published context window of 1M, which determines how much text it can process in a single interaction.

Last updated: July 23, 2026 · Runtime metrics stay blank until BenchLM has a sourced snapshot.

Choose with this week’s evidence

Join 2,000+ readers for ranking moves, new releases, pricing changes, and the evidence behind them.

Free. One email per week.