Model comparison
Qwen3.6-35B-A3B vs Qwen3.7 Plus
Head-to-head evidence from 43 shared benchmark results across 7 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Qwen3.6-35B-A3B #104 (Estimated); Qwen3.7 Plus #21 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Qwen3.6-35B-A3B and Qwen3.7 Plus share 43 comparable benchmark results. 5 of 8 categories are comparable. 14 results are unique to Qwen3.6-35B-A3B; 26 to Qwen3.7 Plus.
Updated July 23, 2026- Shared results
- 43
- Qwen3.6-35B-A3B only
- 14
- Qwen3.7 Plus only
- 26
- Comparable categories
- 5 / 8
Pick Qwen3.7 Plus 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 43 shared benchmark results across 7 evidence categories; 5 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 Plus is clearly ahead on the BenchAlign aggregate, 67.22 to 51.47. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Qwen3.7 Plus's sharpest advantage is in agentic, where it averages 71.7 against 51.5. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 51.5% to 70.3%.
Qwen3.7 Plus 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 | Qwen3.6-35B-A3B | Δ | Qwen3.7 Plus |
|---|---|---|---|
| Agentic | Qwen3.6-35B-A3B51.5 | Margin→ 20.2 | Qwen3.7 Plus71.7 |
| Knowledge | Qwen3.6-35B-A3B51.4 | Margin→ 8.7 | Qwen3.7 Plus60.1 |
| Multimodal | Qwen3.6-35B-A3B76.3 | Margin→ 5.2 | Qwen3.7 Plus81.5 |
| Math | Qwen3.6-35B-A3B88.2 | Margin→ 4.7 | Qwen3.7 Plus92.9 |
| Coding | Qwen3.6-35B-A3B73.8 | Margin→ 1.8 | Qwen3.7 Plus75.6 |
| Reasoning | Qwen3.6-35B-A3BNot measured | MarginNo overlap | Qwen3.7 Plus91.7 |
| Multilingual | Qwen3.6-35B-A3BNot measured | MarginNo overlap | Qwen3.7 Plus85.4 |
| Inst. Following | Qwen3.6-35B-A3BNot measured | MarginNo overlap | Qwen3.7 Plus84.5 |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
Terminal-Bench 2.0
AgenticA 51.5%B 70.3%Winner: Qwen3.7 PlusΔ 18.8Terminal-Bench 2.0: Qwen3.6-35B-A3B scored 51.5%; Qwen3.7 Plus scored 70.3%. Qwen3.7 Plus wins this benchmark. - Source ↗
HLE
KnowledgeA 21.4%B 34.7%Winner: Qwen3.7 PlusΔ 13.3HLE: Qwen3.6-35B-A3B scored 21.4%; Qwen3.7 Plus scored 34.7%. Qwen3.7 Plus wins this benchmark. - Source ↗
HMMT Feb 2026
MathA 83.6%B 92.9%Winner: Qwen3.7 PlusΔ 9.3HMMT Feb 2026: Qwen3.6-35B-A3B scored 83.6%; Qwen3.7 Plus scored 92.9%. Qwen3.7 Plus wins this benchmark. - Source ↗
LiveCodeBench
CodingA 80.4%B 89.6%Winner: Qwen3.7 PlusΔ 9.2LiveCodeBench: Qwen3.6-35B-A3B scored 80.4%; Qwen3.7 Plus scored 89.6%. Qwen3.7 Plus wins this benchmark. - Source ↗
SWE-bench Pro
CodingA 49.5%B 57.6%Winner: Qwen3.7 PlusΔ 8.1SWE-bench Pro: Qwen3.6-35B-A3B scored 49.5%; Qwen3.7 Plus scored 57.6%. Qwen3.7 Plus wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Qwen3.6-35B-A3B | Qwen3.7 Plus | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Qwen3.6-35B-A3BNot available | Qwen3.7 PlusNot available | A complete price comparison is not available. |
| Generation speedtokens per second | Qwen3.6-35B-A3BNot available | Qwen3.7 PlusNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Qwen3.6-35B-A3BNot available | Qwen3.7 PlusNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Qwen3.6-35B-A3B262K | Qwen3.7 Plus1M | Qwen3.7 Plus lists the larger context window. |
Benchmark Deep Dive
AgenticQwen3.7 Plus wins20 benchmarks
| Benchmark | Qwen3.6-35B-A3B | Qwen3.7 Plus | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 51.5% | 70.3% | Qwen3.7 Plus leads |
| Claw-EvalSource | 68.7% | 62.7% | Qwen3.6-35B-A3B leads |
| QwenClawBenchSource | 52.6% | 61.8% | Qwen3.7 Plus leads |
| QwenWebBenchSource | 1397 | 1536 | Qwen3.7 Plus leads |
| τ³-bench resultsSource | 67.2% | — | Not comparable |
| VITA-BenchSource | 35.6% | 45.6% | Qwen3.7 Plus leads |
| DeepPlanningSource | 25.9% | 62.3% | Qwen3.7 Plus leads |
| ToolathlonSource | 26.9% | — | Not comparable |
| MCP AtlasSource | 62.8% | 73.2% | Qwen3.7 Plus leads |
| WideResearchSource | 60.1% | — | Not comparable |
| AA Agentic IndexSource | 21.4% | 20.8% | Qwen3.6-35B-A3B leads |
| τ²-bench resultsSource | 95.3% | 93% | Qwen3.6-35B-A3B leads |
| GDPval-AASource | 27.4% | 21.8% | Qwen3.6-35B-A3B leads |
| GDPval-AASource | 1049 | 936 | Qwen3.6-35B-A3B leads |
| Gert LabsSource | 42.65% | — | Not comparable |
| BFCL v4Source | — | 72.9% | Not comparable |
| OSWorld-VerifiedSource | — | 73.3% | Not comparable |
| AndroidWorldSource | — | 81.0% | Not comparable |
| APEX-Agents-AASource | — | 22.4% | Not comparable |
| OSWorld 2.0Source | — | 2.8% | Not comparable |
CodingQwen3.7 Plus wins9 benchmarks
| Benchmark | Qwen3.6-35B-A3B | Qwen3.7 Plus | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 73.4% | 77.7% | Qwen3.7 Plus leads |
| SWE MultilingualSource | 67.2% | 75.8% | Qwen3.7 Plus leads |
| SWE-bench ProSource | 49.5% | 57.6% | Qwen3.7 Plus leads |
| Terminal-Bench 2.0Source | 51.5% | 70.3% | Qwen3.7 Plus leads |
| LiveCodeBenchSource | 80.4% | 89.6% | Qwen3.7 Plus leads |
| NL2RepoSource | 29.4% | 41.1% | Qwen3.7 Plus leads |
| AA Coding IndexSource | 41.9% | 55.9% | Qwen3.7 Plus leads |
| AA-SciCodeSource | 35.8% | 45.5% | Qwen3.7 Plus leads |
| SciCodeSource | — | 51.3% | Not comparable |
Reasoning3 benchmarks
KnowledgeQwen3.7 Plus wins14 benchmarks
| Benchmark | Qwen3.6-35B-A3B | Qwen3.7 Plus | Result |
|---|---|---|---|
| MMLU-ProSource | 85.2% | 88.5% | Qwen3.7 Plus leads |
| SuperGPQASource | 64.7% | 71.4% | Qwen3.7 Plus leads |
| C-EvalSource | 90% | — | Not comparable |
| GPQASource | 86% | 90.3% | Qwen3.7 Plus leads |
| HLESource | 21.4% | 34.7% | Qwen3.7 Plus leads |
| Artificial Analysis Intelligence IndexSource | 31.6% | 39.0% | Qwen3.7 Plus leads |
| AA-GPQA DiamondSource | 84.1% | 90.0% | Qwen3.7 Plus leads |
| AA-HLESource | 20.2% | 33.4% | Qwen3.7 Plus leads |
| AA-Omniscience IndexSource | -21.4% | 2.4% | Qwen3.7 Plus leads |
| AA-Omniscience AccuracySource | 18.9% | 22.2% | Qwen3.7 Plus leads |
| AA-Omniscience Hallucination RateSource | 49.7% | 25.5% | Qwen3.7 Plus leads |
| GPQA-DSource | — | 90.3% | Not comparable |
| MMLU-ReduxSource | — | 94.5% | Not comparable |
| MMMLUSource | — | 89.0% | Not comparable |
MathQwen3.7 Plus wins7 benchmarks
| Benchmark | Qwen3.6-35B-A3B | Qwen3.7 Plus | Result |
|---|---|---|---|
| HMMT Feb 2025Source | 90.7% | — | Not comparable |
| HMMT Nov 2025Source | 89.1% | — | Not comparable |
| HMMT Feb 2026Source | 83.6% | 92.9% | Qwen3.7 Plus leads |
| MMAnswerBenchSource | 78.9% | — | Not comparable |
| AIME26Source | 92.7% | — | Not comparable |
| IMOAnswerBenchSource | — | 86.0% | Not comparable |
| ApexSource | — | 22.7% | Not comparable |
Multilingual5 benchmarks
MultimodalQwen3.7 Plus wins22 benchmarks
| Benchmark | Qwen3.6-35B-A3B | Qwen3.7 Plus | Result |
|---|---|---|---|
| MMMUSource | 81.7% | — | Not comparable |
| MMMU-ProSource | 75.3% | 79% | Qwen3.7 Plus leads |
| RealWorldQASource | 85.3% | 86.9% | Qwen3.7 Plus leads |
| OmniDocBench 1.5Source | 89.9% | 91.4% | Qwen3.7 Plus leads |
| CharXivSource | 78% | 85.9% | Qwen3.7 Plus leads |
| SimpleVQASource | 58.9% | 81.7% | Qwen3.7 Plus leads |
| CC-OCRSource | 81.9% | — | Not comparable |
| AI2D_TESTSource | 92.7% | — | Not comparable |
| RefCOCO (avg)Source | 92.0% | — | Not comparable |
| ODINW13Source | 50.8% | 51.1% | Qwen3.7 Plus leads |
| Video-MME (with subtitle)Source | 86.6% | 88.0% | Qwen3.7 Plus leads |
| Video-MME (w/o subtitle)Source | 82.5% | — | Not comparable |
| VideoMMMUSource | 83.7% | 85.4% | Qwen3.7 Plus leads |
| MLVU (M-Avg)Source | 86.2% | 87.4% | Qwen3.7 Plus leads |
| AA-MMMU-ProSource | 75.0% | 80.5% | Qwen3.7 Plus leads |
| MathVisionSource | — | 90.3% | Not comparable |
| ERQASource | — | 69.8% | Not comparable |
| MedXpertQA (MM)Source | — | 71.0% | Not comparable |
| ScreenSpot ProSource | — | 79.0% | Not comparable |
| MMSearch-PlusSource | — | 41.4% | Not comparable |
| OCRBench V2Source | — | 70.7% | Not comparable |
| Design Arena WebsiteSource | — | 1288 | Not comparable |
Frequently Asked Questions (6)
Which is better, Qwen3.6-35B-A3B or Qwen3.7 Plus?
Qwen3.7 Plus is ahead on BenchLM's BenchAlign leaderboard, 67.22 to 51.47. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 51.5% and 70.3%.
Which is better for knowledge tasks, Qwen3.6-35B-A3B or Qwen3.7 Plus?
Qwen3.7 Plus has the edge for knowledge tasks in this comparison, averaging 60.1 versus 51.4. Inside this category, AA-Omniscience Hallucination Rate is the benchmark that creates the most daylight between them.
Which is better for coding, Qwen3.6-35B-A3B or Qwen3.7 Plus?
Qwen3.7 Plus has the edge for coding in this comparison, averaging 75.6 versus 73.8. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.
Which is better for math, Qwen3.6-35B-A3B or Qwen3.7 Plus?
Qwen3.7 Plus has the edge for math in this comparison, averaging 92.9 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 Plus?
Qwen3.7 Plus has the edge for agentic tasks in this comparison, averaging 71.7 versus 51.5. Inside this category, QwenWebBench is the benchmark that creates the most daylight between them.
Which is better for multimodal and grounded tasks, Qwen3.6-35B-A3B or Qwen3.7 Plus?
Qwen3.7 Plus has the edge for multimodal and grounded tasks in this comparison, averaging 81.5 versus 76.3. Inside this category, SimpleVQA is the benchmark that creates the most daylight between them.
Related Comparisons
Explore More
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.