Skip to main content
BenchLM
Advertise

Model comparison

Qwen3.6-35B-A3B vs Qwen3.7 Max

Data verified

Head-to-head evidence from 33 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

51.47/100
Margin
21.4pts
winning →
72.84/100
0 category wins4 category wins

Public leaderboard positions: Qwen3.6-35B-A3B #104 (Estimated); Qwen3.7 Max #10 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Qwen3.6-35B-A3B and Qwen3.7 Max share 33 comparable benchmark results. 4 of 8 categories are comparable. 24 results are unique to Qwen3.6-35B-A3B; 25 to Qwen3.7 Max.

Updated July 23, 2026
Shared results
33
Qwen3.6-35B-A3B only
24
Qwen3.7 Max only
25
Comparable categories
4 / 8

Pick Qwen3.7 Max if you want the stronger benchmark profile. Qwen3.6-35B-A3B 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 33 shared benchmark results across 6 evidence categories; 4 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 51.47. 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 agentic, where it averages 69.7 against 51.5. The single biggest benchmark swing on the page is HLE, 21.4% to 41.4%.

Qwen3.7 Max gives you the larger context window at 1M, compared with 262K for Qwen3.6-35B-A3B.

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 Qwen3.6-35B-A3B and Qwen3.7 Max
CategoryQwen3.6-35B-A3BΔQwen3.7 Max
AgenticQwen3.6-35B-A3B51.5Margin→ 18.2Qwen3.7 Max69.7
KnowledgeQwen3.6-35B-A3B51.4Margin→ 12.8Qwen3.7 Max64.2
MathQwen3.6-35B-A3B88.2Margin→ 8.9Qwen3.7 Max97.1
CodingQwen3.6-35B-A3B73.8Margin→ 4.1Qwen3.7 Max77.9
ReasoningQwen3.6-35B-A3BNot measuredMarginNo overlapQwen3.7 Max90.4
MultilingualQwen3.6-35B-A3BNot measuredMarginNo overlapQwen3.7 Max87.0
MultimodalQwen3.6-35B-A3B76.3MarginNo overlapQwen3.7 MaxNot measured
Inst. FollowingQwen3.6-35B-A3BNot measuredMarginNo overlapQwen3.7 Max84.4

Decisive benchmark drivers

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

More
A · Qwen3.6-35B-A3BB · Qwen3.7 Max

  1. HLE

    Knowledge
    Source ↗
    A 21.4%B 41.4%
    Winner: Qwen3.7 MaxΔ 20
    HLE: Qwen3.6-35B-A3B scored 21.4%; Qwen3.7 Max scored 41.4%. Qwen3.7 Max wins this benchmark.

  2. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 51.5%B 69.7%
    Winner: Qwen3.7 MaxΔ 18.2
    Terminal-Bench 2.0: Qwen3.6-35B-A3B scored 51.5%; Qwen3.7 Max scored 69.7%. Qwen3.7 Max wins this benchmark.

  3. HMMT Feb 2026

    Math
    Source ↗
    A 83.6%B 97.1%
    Winner: Qwen3.7 MaxΔ 13.5
    HMMT Feb 2026: Qwen3.6-35B-A3B scored 83.6%; Qwen3.7 Max scored 97.1%. Qwen3.7 Max wins this benchmark.

  4. LiveCodeBench

    Coding
    Source ↗
    A 80.4%B 91.6%
    Winner: Qwen3.7 MaxΔ 11.2
    LiveCodeBench: Qwen3.6-35B-A3B scored 80.4%; Qwen3.7 Max scored 91.6%. Qwen3.7 Max wins this benchmark.

  5. SWE-bench Pro

    Coding
    Source ↗
    A 49.5%B 60.6%
    Winner: Qwen3.7 MaxΔ 11.1
    SWE-bench Pro: Qwen3.6-35B-A3B scored 49.5%; Qwen3.7 Max scored 60.6%. Qwen3.7 Max wins this benchmark.

Operational comparison

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

MetricQwen3.6-35B-A3BQwen3.7 MaxComparison
Input / output priceUSD per 1M tokensQwen3.6-35B-A3BNot availableQwen3.7 MaxNot availableA complete price comparison is not available.
Generation speedtokens per secondQwen3.6-35B-A3BNot availableQwen3.7 MaxNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenQwen3.6-35B-A3BNot availableQwen3.7 MaxNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensQwen3.6-35B-A3B262KQwen3.7 Max1MQwen3.7 Max lists the larger context window.

Benchmark Deep Dive

AgenticQwen3.7 Max wins
BenchmarkQwen3.6-35B-A3BQwen3.7 MaxResult
Terminal-Bench 2.0Source 51.5%69.7%Qwen3.7 Max leads
Claw-EvalSource 68.7%65.2%Qwen3.6-35B-A3B leads
QwenClawBenchSource 52.6%64.3%Qwen3.7 Max leads
QwenWebBenchSource 13971568Qwen3.7 Max leads
τ³-bench resultsSource 67.2%—Not comparable
VITA-BenchSource 35.6%47.9%Qwen3.7 Max leads
DeepPlanningSource 25.9%—Not comparable
ToolathlonSource 26.9%—Not comparable
MCP AtlasSource 62.8%76.4%Qwen3.7 Max leads
WideResearchSource 60.1%—Not comparable
AA Agentic IndexSource 21.4%30.6%Qwen3.7 Max leads
τ²-bench resultsSource 95.3%94.7%Qwen3.6-35B-A3B leads
GDPval-AASource 27.4%38.7%Qwen3.7 Max leads
GDPval-AASource 10491273Qwen3.7 Max leads
Gert LabsSource 42.65%64.27%Qwen3.7 Max leads
BFCL v4Source —75.0%Not comparable
HLE w/ toolsSource —53.5%Not comparable
ResearchClawBenchSource —18.7%Not comparable
AA BriefcaseSource —908Not 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
CodingQwen3.7 Max wins
BenchmarkQwen3.6-35B-A3BQwen3.7 MaxResult
SWE-bench VerifiedSource 73.4%80.4%Qwen3.7 Max leads
SWE MultilingualSource 67.2%78.3%Qwen3.7 Max leads
SWE-bench ProSource 49.5%60.6%Qwen3.7 Max leads
Terminal-Bench 2.0Source 51.5%69.7%Qwen3.7 Max leads
LiveCodeBenchSource 80.4%91.6%Qwen3.7 Max leads
NL2RepoSource 29.4%47.2%Qwen3.7 Max leads
AA Coding IndexSource 41.9%66.0%Qwen3.7 Max leads
AA-SciCodeSource 35.8%48.8%Qwen3.7 Max leads
SciCodeSource —53.5%Not comparable
Reasoning
BenchmarkQwen3.6-35B-A3BQwen3.7 MaxResult
AA-LCRSource 63.7%69.0%Qwen3.7 Max leads
CritPtSource 0.3%13.4%Qwen3.7 Max leads
MRCRv2Source —90.4%Not comparable
KnowledgeQwen3.7 Max wins
BenchmarkQwen3.6-35B-A3BQwen3.7 MaxResult
MMLU-ProSource 85.2%89.6%Qwen3.7 Max leads
SuperGPQASource 64.7%73.6%Qwen3.7 Max leads
C-EvalSource 90%—Not comparable
GPQASource 86%92.4%Qwen3.7 Max leads
HLESource 21.4%41.4%Qwen3.7 Max leads
Artificial Analysis Intelligence IndexSource 31.6%46.0%Qwen3.7 Max leads
AA-GPQA DiamondSource 84.1%92.3%Qwen3.7 Max leads
AA-HLESource 20.2%38.1%Qwen3.7 Max leads
AA-Omniscience IndexSource -21.4%14.1%Qwen3.7 Max leads
AA-Omniscience AccuracySource 18.9%30.1%Qwen3.7 Max leads
AA-Omniscience Hallucination RateSource 49.7%22.9%Qwen3.7 Max leads
GPQA-DSource —92.4%Not comparable
MMLU-ReduxSource —95%Not comparable
MMMLUSource —90.3%Not comparable
MathQwen3.7 Max wins
BenchmarkQwen3.6-35B-A3BQwen3.7 MaxResult
HMMT Feb 2025Source 90.7%—Not comparable
HMMT Nov 2025Source 89.1%—Not comparable
HMMT Feb 2026Source 83.6%97.1%Qwen3.7 Max leads
MMAnswerBenchSource 78.9%—Not comparable
AIME26Source 92.7%—Not comparable
IMOAnswerBenchSource —90.0%Not comparable
ApexSource —44.5%Not comparable
Multilingual
BenchmarkQwen3.6-35B-A3BQwen3.7 MaxResult
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
BenchmarkQwen3.6-35B-A3BQwen3.7 MaxResult
MMMUSource 81.7%—Not comparable
MMMU-ProSource 75.3%—Not comparable
RealWorldQASource 85.3%—Not comparable
OmniDocBench 1.5Source 89.9%—Not comparable
CharXivSource 78%—Not comparable
SimpleVQASource 58.9%—Not comparable
CC-OCRSource 81.9%—Not comparable
AI2D_TESTSource 92.7%—Not comparable
RefCOCO (avg)Source 92.0%—Not comparable
ODINW13Source 50.8%—Not comparable
Video-MME (with subtitle)Source 86.6%—Not comparable
Video-MME (w/o subtitle)Source 82.5%—Not comparable
VideoMMMUSource 83.7%—Not comparable
MLVU (M-Avg)Source 86.2%—Not comparable
AA-MMMU-ProSource 75.0%—Not comparable
Design Arena WebsiteSource —1293Not comparable
Inst. Following
BenchmarkQwen3.6-35B-A3BQwen3.7 MaxResult
AA-IFBenchSource 64.4%80.5%Qwen3.7 Max leads
IFEvalSource —94.3%Not comparable
IFBenchSource —79.1%Not comparable
Frequently Asked Questions (5)

Which is better, Qwen3.6-35B-A3B or Qwen3.7 Max?

Qwen3.7 Max is ahead on BenchLM's BenchAlign leaderboard, 72.84 to 51.47. The biggest single separator in this matchup is HLE, where the scores are 21.4% and 41.4%.

Which is better for knowledge tasks, Qwen3.6-35B-A3B or Qwen3.7 Max?

Qwen3.7 Max has the edge for knowledge tasks in this comparison, averaging 64.2 versus 51.4. Inside this category, AA-Omniscience Index is the benchmark that creates the most daylight between them.

Which is better for coding, Qwen3.6-35B-A3B or Qwen3.7 Max?

Qwen3.7 Max has the edge for coding in this comparison, averaging 77.9 versus 73.8. Inside this category, AA Coding Index is the benchmark that creates the most daylight between them.

Which is better for math, Qwen3.6-35B-A3B or Qwen3.7 Max?

Qwen3.7 Max has the edge for math in this comparison, averaging 97.1 versus 88.2. Inside this category, HMMT Feb 2026 is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Qwen3.6-35B-A3B or Qwen3.7 Max?

Qwen3.7 Max has the edge for agentic tasks in this comparison, averaging 69.7 versus 51.5. 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 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.