Skip to main content

Model comparison

Qwen3.6 Plus vs Step 3.7 Flash

Data verified

Head-to-head evidence from 24 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

65.05/100
Margin
14.3pts
← winning
50.76/100
1 category wins1 category wins

Verified leaderboard positions: Qwen3.6 Plus #14; Step 3.7 Flash unranked

BenchAlign evidence: Qwen3.6 Plus supported; Step 3.7 Flash estimated. Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Qwen3.6 Plus and Step 3.7 Flash share 24 comparable benchmark results. 2 of 8 categories are comparable. 37 results are unique to Qwen3.6 Plus; 6 to Step 3.7 Flash.

Updated July 16, 2026
Shared results
24
Qwen3.6 Plus only
37
Step 3.7 Flash only
6
Comparable categories
2 / 8

Pick Qwen3.6 Plus if you want the stronger benchmark profile. Step 3.7 Flash only becomes the better choice if agentic is the priority.

Confidence note. This is a partial-evidence comparison with 24 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

Qwen3.6 Plus is clearly ahead on the provisional aggregate, 63 to 57. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Qwen3.6 Plus's sharpest advantage is in coding, where it averages 70.3 against 56.3. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 61.6% to 59.5%. Step 3.7 Flash does hit back in agentic, so the answer changes if that is the part of the workload you care about most.

Qwen3.6 Plus 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 Qwen3.6 Plus and Step 3.7 Flash
CategoryQwen3.6 PlusΔStep 3.7 Flash
CodingQwen3.6 Plus70.3Margin 14.0Step 3.7 Flash56.3
AgenticQwen3.6 Plus61.6Margin 4.8Step 3.7 Flash66.4
ReasoningQwen3.6 Plus62.0MarginNo overlapStep 3.7 FlashNot measured
KnowledgeQwen3.6 Plus57.5MarginNo overlapStep 3.7 FlashNot measured
MathQwen3.6 Plus60.5MarginNo overlapStep 3.7 FlashNot measured
MultilingualQwen3.6 Plus84.7MarginNo overlapStep 3.7 FlashNot measured
MultimodalQwen3.6 Plus79.8MarginNo overlapStep 3.7 FlashNot measured
Inst. FollowingQwen3.6 Plus82.3MarginNo overlapStep 3.7 FlashNot measured

Decisive benchmark drivers

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

More
A · Qwen3.6 PlusB · Step 3.7 Flash
  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 61.6%B 59.5%
    Winner: Qwen3.6 PlusΔ 2.1
    Terminal-Bench 2.0: Qwen3.6 Plus scored 61.6%; Step 3.7 Flash scored 59.5%. Qwen3.6 Plus wins this benchmark.
  2. SWE-bench Pro

    Coding
    Source ↗
    A 56.6%B 56.3%
    Winner: Qwen3.6 PlusΔ 0.3
    SWE-bench Pro: Qwen3.6 Plus scored 56.6%; Step 3.7 Flash scored 56.3%. Qwen3.6 Plus wins this benchmark.

Operational comparison

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

MetricQwen3.6 PlusStep 3.7 FlashComparison
Input / output priceUSD per 1M tokensQwen3.6 PlusNot availableStep 3.7 Flash$0.2 input / $1.15 outputA complete price comparison is not available.
Generation speedtokens per secondQwen3.6 PlusNot availableStep 3.7 FlashNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenQwen3.6 PlusNot availableStep 3.7 FlashNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensQwen3.6 Plus1MStep 3.7 Flash256KQwen3.6 Plus lists the larger context window.

Benchmark Deep Dive

AgenticStep 3.7 Flash wins
BenchmarkQwen3.6 PlusStep 3.7 FlashResult
Terminal-Bench 2.0Source 61.6%59.5%Qwen3.6 Plus leads
Claw-EvalSource 58.8%67.1%Step 3.7 Flash leads
QwenClawBenchSource 57.2%Not comparable
τ³-bench resultsSource 70.7%Not comparable
VITA-BenchSource 44.3%Not comparable
DeepPlanningSource 41.5%Not comparable
ToolathlonSource 39.8%49.5%Step 3.7 Flash leads
MCP AtlasSource 48.2%Not comparable
MCP-TasksSource 74.1%Not comparable
WideResearchSource 74.3%Not comparable
AA Agentic IndexSource 27.6%21.5%Qwen3.6 Plus leads
τ²-bench resultsSource 97.7%98.5%Step 3.7 Flash leads
GDPval-AASource 31.8%25.9%Qwen3.6 Plus leads
GDPval-AASource 11351017Qwen3.6 Plus leads
Gert LabsSource 50.60%51.57%Step 3.7 Flash leads
ResearchClawBenchSource 18.0%Not comparable
BrowseCompSource 75.8%Not comparable
DeepSearchQASource 92.8%Not comparable
HLE w/ toolsSource 47.2%Not comparable
APEX-Agents-AASource 14.8%Not comparable
CodingQwen3.6 Plus wins
BenchmarkQwen3.6 PlusStep 3.7 FlashResult
SWE-bench VerifiedSource 78.8%Not comparable
SWE-bench ProSource 56.6%56.3%Qwen3.6 Plus leads
SWE MultilingualSource 73.8%Not comparable
LiveCodeBench v6Source 87.1%Not comparable
Vibe Code BenchSource 25.56%Not comparable
AA Coding IndexSource 54.5%39.6%Qwen3.6 Plus leads
Terminal-Bench HardSource 43.9%35.6%Qwen3.6 Plus leads
AA-SciCodeSource 40.7%40.0%Qwen3.6 Plus leads
Terminal-Bench 2.0Source 59.5%Not comparable
Reasoning
BenchmarkQwen3.6 PlusStep 3.7 FlashResult
AI-NeedleSource 68.3%Not comparable
LongBench v2Source 62%Not comparable
AA-LCRSource 69.7%63.7%Qwen3.6 Plus leads
CritPtSource 2.9%2.3%Qwen3.6 Plus leads
Knowledge
BenchmarkQwen3.6 PlusStep 3.7 FlashResult
GPQASource 90.4%Not comparable
SuperGPQASource 71.6%Not comparable
MMLU-ProSource 88.5%Not comparable
MMLU-ReduxSource 94.5%Not comparable
C-EvalSource 93.3%Not comparable
HLESource 28.8%Not comparable
Artificial Analysis Intelligence IndexSource 39.6%30.3%Qwen3.6 Plus leads
AA-GPQA DiamondSource 88.2%80.9%Qwen3.6 Plus leads
AA-HLESource 25.7%19.9%Qwen3.6 Plus leads
AA-Omniscience IndexSource 2.7%-37.5%Qwen3.6 Plus leads
AA-Omniscience AccuracySource 26.2%25.4%Qwen3.6 Plus leads
AA-Omniscience Hallucination RateSource 32.0%84.4%Qwen3.6 Plus leads
Math
BenchmarkQwen3.6 PlusStep 3.7 FlashResult
AIME26Source 95.3%Not comparable
HMMT Feb 2025Source 96.7%Not comparable
HMMT Nov 2025Source 94.6%Not comparable
HMMT Feb 2026Source 87.8%Not comparable
MMAnswerBenchSource 83.8%Not comparable
FrontierMath v2 (Tiers 1-3)Source 26.207%Not comparable
FrontierMath v2 (Tier 4)Source 8.333%Not comparable
Multilingual
BenchmarkQwen3.6 PlusStep 3.7 FlashResult
MMLU-ProXSource 84.7%Not comparable
NOVA-63Source 57.9%Not comparable
Multimodal
BenchmarkQwen3.6 PlusStep 3.7 FlashResult
MMMUSource 86.0%Not comparable
MMMU-ProSource 78.8%Not comparable
MathVisionSource 88.0%Not comparable
VideoMMMUSource 84.0%Not comparable
ScreenSpot ProSource 68.2%Not comparable
CharXivSource 81.5%Not comparable
V*Source 96.9%95.3%Qwen3.6 Plus leads
AA-MMMU-ProSource 78.0%75.3%Qwen3.6 Plus leads
Design Arena WebsiteSource 12541218Qwen3.6 Plus leads
SimpleVQASource 79.2%Not comparable
Inst. Following
BenchmarkQwen3.6 PlusStep 3.7 FlashResult
IFEvalSource 94.3%Not comparable
IFBenchSource 75.8%Not comparable
AA-IFBenchSource 75.2%67.3%Qwen3.6 Plus leads
Frequently Asked Questions (3)

Which is better, Qwen3.6 Plus or Step 3.7 Flash?

Qwen3.6 Plus is ahead on BenchLM's provisional leaderboard, 63 to 57. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 61.6% and 59.5%.

Which is better for coding, Qwen3.6 Plus or Step 3.7 Flash?

Qwen3.6 Plus has the edge for coding in this comparison, averaging 70.3 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, Qwen3.6 Plus or Step 3.7 Flash?

Step 3.7 Flash has the edge for agentic tasks in this comparison, averaging 66.4 versus 61.6. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.

Related Comparisons

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.