Skip to main content
BenchLM
Advertise

Model comparison

Qwen3.7 Max vs Step 3.7 Flash

Data verified

Head-to-head evidence from 23 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

60.99/100
Margin
10.2pts
← winning
50.76/100
2 category wins0 category wins

Verified leaderboard positions: Qwen3.7 Max #2; Step 3.7 Flash unranked

BenchAlign evidence: Qwen3.7 Max estimated; Step 3.7 Flash estimated. Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Qwen3.7 Max and Step 3.7 Flash share 23 comparable benchmark results. 2 of 8 categories are comparable. 34 results are unique to Qwen3.7 Max; 7 to Step 3.7 Flash.

Updated July 16, 2026
Shared results
23
Qwen3.7 Max only
34
Step 3.7 Flash only
7
Comparable categories
2 / 8

Pick Qwen3.7 Max if you want the stronger benchmark profile. Step 3.7 Flash only becomes the better choice if its workflow or ecosystem matters more than the raw scoreboard.

Confidence note. This is a partial-evidence comparison with 23 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

Qwen3.7 Max is clearly ahead on the provisional aggregate, 83 to 57. 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 coding, where it averages 77.9 against 56.3. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 69.7% to 59.5%.

Qwen3.7 Max 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 Qwen3.7 Max and Step 3.7 Flash
CategoryQwen3.7 MaxΔStep 3.7 Flash
CodingQwen3.7 Max77.9Margin← 21.6Step 3.7 Flash56.3
AgenticQwen3.7 Max69.7Margin← 3.3Step 3.7 Flash66.4
ReasoningQwen3.7 Max90.4MarginNo overlapStep 3.7 FlashNot measured
KnowledgeQwen3.7 Max64.5MarginNo overlapStep 3.7 FlashNot measured
MathQwen3.7 Max97.1MarginNo overlapStep 3.7 FlashNot measured
MultilingualQwen3.7 Max87.0MarginNo overlapStep 3.7 FlashNot measured
Inst. FollowingQwen3.7 Max84.4MarginNo overlapStep 3.7 FlashNot measured

Decisive benchmark drivers

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

More
A · Qwen3.7 MaxB · Step 3.7 Flash

  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 69.7%B 59.5%
    Winner: Qwen3.7 MaxΔ 10.2
    Terminal-Bench 2.0: Qwen3.7 Max scored 69.7%; Step 3.7 Flash scored 59.5%. Qwen3.7 Max wins this benchmark.

  2. SWE-bench Pro

    Coding
    Source ↗
    A 60.6%B 56.3%
    Winner: Qwen3.7 MaxΔ 4.3
    SWE-bench Pro: Qwen3.7 Max scored 60.6%; Step 3.7 Flash scored 56.3%. 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 MaxStep 3.7 FlashComparison
Input / output priceUSD per 1M tokensQwen3.7 MaxNot availableStep 3.7 Flash$0.2 input / $1.15 outputA complete price comparison is not available.
Generation speedtokens per secondQwen3.7 MaxNot availableStep 3.7 FlashNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenQwen3.7 MaxNot availableStep 3.7 FlashNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensQwen3.7 Max1MStep 3.7 Flash256KQwen3.7 Max lists the larger context window.

Benchmark Deep Dive

AgenticQwen3.7 Max wins
BenchmarkQwen3.7 MaxStep 3.7 FlashResult
Terminal-Bench 2.0Source 69.7%59.5%Qwen3.7 Max leads
QwenClawBenchSource 64.3%—Not comparable
QwenWebBenchSource 1568—Not comparable
Claw-EvalSource 65.2%67.1%Step 3.7 Flash leads
BFCL v4Source 75.0%—Not comparable
MCP AtlasSource 76.4%—Not comparable
VITA-BenchSource 47.9%—Not comparable
HLE w/ toolsSource 53.5%47.2%Qwen3.7 Max leads
AA Agentic IndexSource 30.6%21.5%Qwen3.7 Max leads
τ²-bench resultsSource 94.7%98.5%Step 3.7 Flash leads
GDPval-AASource 38.7%25.9%Qwen3.7 Max leads
GDPval-AASource 12731017Qwen3.7 Max leads
Gert LabsSource 64.27%51.57%Qwen3.7 Max leads
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
BrowseCompSource —75.8%Not comparable
DeepSearchQASource —92.8%Not comparable
ToolathlonSource —49.5%Not comparable
APEX-Agents-AASource —14.8%Not comparable
CodingQwen3.7 Max wins
BenchmarkQwen3.7 MaxStep 3.7 FlashResult
SWE-bench VerifiedSource 80.4%—Not comparable
SWE-bench ProSource 60.6%56.3%Qwen3.7 Max leads
SWE MultilingualSource 78.3%—Not comparable
NL2RepoSource 47.2%—Not comparable
SciCodeSource 53.5%—Not comparable
LiveCodeBenchSource 91.6%—Not comparable
Terminal-Bench 2.0Source 69.7%59.5%Qwen3.7 Max leads
AA Coding IndexSource 66.0%39.6%Qwen3.7 Max leads
Terminal-Bench HardSource 50.8%35.6%Qwen3.7 Max leads
AA-SciCodeSource 48.8%40.0%Qwen3.7 Max leads
AA Terminal-Bench 2.1Source 74.5%—Not comparable
Reasoning
BenchmarkQwen3.7 MaxStep 3.7 FlashResult
MRCRv2Source 90.4%—Not comparable
CritPtSource 13.4%2.3%Qwen3.7 Max leads
AA-LCRSource 69.0%63.7%Qwen3.7 Max leads
Knowledge
BenchmarkQwen3.7 MaxStep 3.7 FlashResult
GPQASource 92.4%—Not comparable
GPQA-DSource 92.4%—Not comparable
HLESource 41.4%—Not comparable
MMLU-ProSource 89.6%—Not comparable
MMLU-ReduxSource 95%—Not comparable
SuperGPQASource 73.6%—Not comparable
MMMLUSource 90.3%—Not comparable
Artificial Analysis Intelligence IndexSource 46.0%30.3%Qwen3.7 Max leads
AA-GPQA DiamondSource 92.3%80.9%Qwen3.7 Max leads
AA-HLESource 38.1%19.9%Qwen3.7 Max leads
AA-Omniscience IndexSource 14.1%-37.5%Qwen3.7 Max leads
AA-Omniscience AccuracySource 30.1%25.4%Qwen3.7 Max leads
AA-Omniscience Hallucination RateSource 22.9%84.4%Qwen3.7 Max leads
Math
BenchmarkQwen3.7 MaxStep 3.7 FlashResult
HMMT Feb 2026Source 97.1%—Not comparable
IMOAnswerBenchSource 90.0%—Not comparable
ApexSource 44.5%—Not comparable
Multilingual
BenchmarkQwen3.7 MaxStep 3.7 FlashResult
MMLU-ProXSource 87%—Not comparable
NOVA-63Source 59.0%—Not comparable
INCLUDESource 86.2%—Not comparable
MAXIFESource 89.2%—Not comparable
PolyMathSource 86.5%—Not comparable
Multimodal
BenchmarkQwen3.7 MaxStep 3.7 FlashResult
Design Arena WebsiteSource 12961218Qwen3.7 Max leads
SimpleVQASource —79.2%Not comparable
V*Source —95.3%Not comparable
AA-MMMU-ProSource —75.3%Not comparable
Inst. Following
BenchmarkQwen3.7 MaxStep 3.7 FlashResult
IFEvalSource 94.3%—Not comparable
IFBenchSource 79.1%—Not comparable
AA-IFBenchSource 80.5%67.3%Qwen3.7 Max leads
Frequently Asked Questions (3)

Which is better, Qwen3.7 Max or Step 3.7 Flash?

Qwen3.7 Max is ahead on BenchLM's provisional leaderboard, 83 to 57. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 69.7% and 59.5%.

Which is better for coding, Qwen3.7 Max or Step 3.7 Flash?

Qwen3.7 Max has the edge for coding in this comparison, averaging 77.9 versus 56.3. Inside this category, AA Coding Index is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Qwen3.7 Max or Step 3.7 Flash?

Qwen3.7 Max has the edge for agentic tasks in this comparison, averaging 69.7 versus 66.4. Inside this category, GDPval-AA 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 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.