Skip to main content

Model comparison

MiniMax M3 vs Qwen3.6-35B-A3B

Data verified

Head-to-head evidence from 28 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

MiniMax
69.8/100
Margin
18.4pts
← winning
51.36/100
1 category wins3 category wins

Verified leaderboard positions: MiniMax M3 #18; Qwen3.6-35B-A3B #31

BenchAlign evidence: MiniMax M3 supported; Qwen3.6-35B-A3B estimated. Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. MiniMax M3 and Qwen3.6-35B-A3B share 28 comparable benchmark results. 4 of 8 categories are comparable. 17 results are unique to MiniMax M3; 30 to Qwen3.6-35B-A3B.

Updated July 16, 2026
Shared results
28
MiniMax M3 only
17
Qwen3.6-35B-A3B only
30
Comparable categories
4 / 8

Pick MiniMax M3 if you want the stronger benchmark profile. Qwen3.6-35B-A3B only becomes the better choice if multimodal & grounded is the priority or you want the stronger reasoning-first profile.

Confidence note. This is a partial-evidence comparison with 28 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

MiniMax M3 is clearly ahead on the provisional aggregate, 70 to 58. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

MiniMax M3's sharpest advantage is in agentic, where it averages 72.3 against 51.5. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 66% to 51.5%. Qwen3.6-35B-A3B does hit back in multimodal & grounded, so the answer changes if that is the part of the workload you care about most.

Qwen3.6-35B-A3B is the reasoning model in the pair, while MiniMax M3 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. MiniMax M3 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 MiniMax M3 and Qwen3.6-35B-A3B
CategoryMiniMax M3ΔQwen3.6-35B-A3B
AgenticMiniMax M372.3Margin 20.8Qwen3.6-35B-A3B51.5
MultimodalMiniMax M364.9Margin 11.4Qwen3.6-35B-A3B76.3
MathMiniMax M385.7Margin 2.5Qwen3.6-35B-A3B88.2
CodingMiniMax M372.2Margin 1.6Qwen3.6-35B-A3B73.8
KnowledgeMiniMax M3Not measuredMarginNo overlapQwen3.6-35B-A3B51.8

Decisive benchmark drivers

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

More
A · MiniMax M3B · Qwen3.6-35B-A3B
  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 66%B 51.5%
    Winner: MiniMax M3Δ 14.5
    Terminal-Bench 2.0: MiniMax M3 scored 66%; Qwen3.6-35B-A3B scored 51.5%. MiniMax M3 wins this benchmark.
  2. SWE-bench Pro

    Coding
    Source ↗
    A 59%B 49.5%
    Winner: MiniMax M3Δ 9.5
    SWE-bench Pro: MiniMax M3 scored 59%; Qwen3.6-35B-A3B scored 49.5%. MiniMax M3 wins this benchmark.
  3. SWE-bench Verified

    Coding
    Source ↗
    A 80.5%B 73.4%
    Winner: MiniMax M3Δ 7.1
    SWE-bench Verified: MiniMax M3 scored 80.5%; Qwen3.6-35B-A3B scored 73.4%. MiniMax M3 wins this benchmark.
  4. MMMU-Pro

    Multimodal
    Source ↗
    A 78.1%B 75.3%
    Winner: MiniMax M3Δ 2.8
    MMMU-Pro: MiniMax M3 scored 78.1%; Qwen3.6-35B-A3B scored 75.3%. MiniMax M3 wins this benchmark.

Operational comparison

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

MetricMiniMax M3Qwen3.6-35B-A3BComparison
Input / output priceUSD per 1M tokensMiniMax M3$0.3 input / $1.2 outputQwen3.6-35B-A3BNot availableA complete price comparison is not available.
Generation speedtokens per secondMiniMax M3Not availableQwen3.6-35B-A3BNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenMiniMax M3Not availableQwen3.6-35B-A3BNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensMiniMax M31MQwen3.6-35B-A3B262KMiniMax M3 lists the larger context window.

Benchmark Deep Dive

AgenticMiniMax M3 wins
BenchmarkMiniMax M3Qwen3.6-35B-A3BResult
Terminal-Bench 2.0Source 66%51.5%MiniMax M3 leads
BrowseCompSource 83.5%Not comparable
OSWorld-VerifiedSource 70.1%Not comparable
MCP AtlasSource 74.2%62.8%MiniMax M3 leads
Claw-EvalSource 74.5%68.7%MiniMax M3 leads
AA Agentic IndexSource 35.4%21.4%MiniMax M3 leads
τ²-bench resultsSource 88.9%95.3%Qwen3.6-35B-A3B leads
GDPval-AASource 44.7%27.4%MiniMax M3 leads
GDPval-AASource 13951049MiniMax M3 leads
GDPval rubricsSource 74.7%Not comparable
BankerToolBenchSource 76.1%Not comparable
ResearchClawBenchSource 19.8%Not comparable
OSWorld 2.0Source 4.6%Not comparable
AA BriefcaseSource 1110Not comparable
AA EnterpriseOps-GymSource 32.1%Not comparable
AA Harvey LABSource 6.7%Not comparable
QwenClawBenchSource 52.6%Not comparable
QwenWebBenchSource 1397Not comparable
τ³-bench resultsSource 67.2%Not comparable
VITA-BenchSource 35.6%Not comparable
DeepPlanningSource 25.9%Not comparable
ToolathlonSource 26.9%Not comparable
WideResearchSource 60.1%Not comparable
Gert LabsSource 42.65%Not comparable
CodingQwen3.6-35B-A3B wins
BenchmarkMiniMax M3Qwen3.6-35B-A3BResult
SWE-bench VerifiedSource 80.5%73.4%MiniMax M3 leads
SWE-bench ProSource 59%49.5%MiniMax M3 leads
Terminal-Bench 2.0Source 66.0%51.5%MiniMax M3 leads
NL2RepoSource 42.1%29.4%MiniMax M3 leads
AA Coding IndexSource 58.6%41.9%MiniMax M3 leads
Terminal-Bench HardSource 42.4%34.8%MiniMax M3 leads
AA-SciCodeSource 45.4%35.8%MiniMax M3 leads
VIBE V2Source 50.1%Not comparable
SVG-BenchSource 63.7%Not comparable
KernelBench HardSource 28.8%Not comparable
AA Terminal-Bench 2.1Source 65.2%Not comparable
SWE MultilingualSource 67.2%Not comparable
LiveCodeBenchSource 80.4%Not comparable
Reasoning
BenchmarkMiniMax M3Qwen3.6-35B-A3BResult
AA-LCRSource 74.0%63.7%MiniMax M3 leads
CritPtSource 3.7%0.3%MiniMax M3 leads
Knowledge
BenchmarkMiniMax M3Qwen3.6-35B-A3BResult
Artificial Analysis Intelligence IndexSource 44.4%31.6%MiniMax M3 leads
AA-GPQA DiamondSource 92.9%84.1%MiniMax M3 leads
AA-HLESource 37.1%20.2%MiniMax M3 leads
AA-Omniscience IndexSource 1.4%-21.4%MiniMax M3 leads
AA-Omniscience AccuracySource 15.0%18.9%Qwen3.6-35B-A3B leads
AA-Omniscience Hallucination RateSource 16.1%49.7%MiniMax M3 leads
AA Openness IndexSource 33.3%Not comparable
MMLU-ProSource 85.2%Not comparable
SuperGPQASource 64.7%Not comparable
C-EvalSource 90%Not comparable
GPQASource 86%Not comparable
HLESource 21.4%Not comparable
MathQwen3.6-35B-A3B wins
BenchmarkMiniMax M3Qwen3.6-35B-A3BResult
USAMO 2026Source 85.7%Not comparable
HMMT Feb 2025Source 90.7%Not comparable
HMMT Nov 2025Source 89.1%Not comparable
HMMT Feb 2026Source 83.6%Not comparable
MMAnswerBenchSource 78.9%Not comparable
AIME26Source 92.7%Not comparable
MultimodalQwen3.6-35B-A3B wins
BenchmarkMiniMax M3Qwen3.6-35B-A3BResult
OfficeQA ProSource 45.1%Not comparable
OmniDocBench 1.5Source 91.6%89.9%MiniMax M3 leads
MMMU-ProSource 78.1%75.3%MiniMax M3 leads
VideoMMMUSource 84.6%83.7%MiniMax M3 leads
Video-MME (with subtitle)Source 85.4%86.6%Qwen3.6-35B-A3B leads
Design Arena WebsiteSource 1294Not comparable
AA-MMMU-ProSource 78.6%75.0%MiniMax M3 leads
MMMUSource 81.7%Not comparable
RealWorldQASource 85.3%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 (w/o subtitle)Source 82.5%Not comparable
MLVU (M-Avg)Source 86.2%Not comparable
Inst. Following
BenchmarkMiniMax M3Qwen3.6-35B-A3BResult
AA-IFBenchSource 82.9%64.4%MiniMax M3 leads
Frequently Asked Questions (5)

Which is better, MiniMax M3 or Qwen3.6-35B-A3B?

MiniMax M3 is ahead on BenchLM's provisional leaderboard, 70 to 58. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 66% and 51.5%.

Which is better for coding, MiniMax M3 or Qwen3.6-35B-A3B?

Qwen3.6-35B-A3B has the edge for coding in this comparison, averaging 73.8 versus 72.2. Inside this category, AA Coding Index is the benchmark that creates the most daylight between them.

Which is better for math, MiniMax M3 or Qwen3.6-35B-A3B?

Qwen3.6-35B-A3B has the edge for math in this comparison, averaging 88.2 versus 85.7. MiniMax M3 stays close enough that the answer can still flip depending on your workload.

Which is better for agentic tasks, MiniMax M3 or Qwen3.6-35B-A3B?

MiniMax M3 has the edge for agentic tasks in this comparison, averaging 72.3 versus 51.5. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.

Which is better for multimodal and grounded tasks, MiniMax M3 or Qwen3.6-35B-A3B?

Qwen3.6-35B-A3B has the edge for multimodal and grounded tasks in this comparison, averaging 76.3 versus 64.9. Inside this category, AA-MMMU-Pro 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.