Model comparison
Qwen3.6-27B vs Qwen3.7 Plus
Head-to-head evidence from 41 shared benchmark results across 7 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Qwen3.6-27B #93 (Estimated); Qwen3.7 Plus #21 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Qwen3.6-27B and Qwen3.7 Plus share 41 comparable benchmark results. 5 of 8 categories are comparable. 13 results are unique to Qwen3.6-27B; 28 to Qwen3.7 Plus.
Updated July 23, 2026- Shared results
- 41
- Qwen3.6-27B only
- 13
- Qwen3.7 Plus only
- 28
- Comparable categories
- 5 / 8
Pick Qwen3.7 Plus if you want the stronger benchmark profile. Qwen3.6-27B only becomes the better choice if coding is the priority.
Confidence note. This is a partial-evidence comparison with 41 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 53.82. 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 59.3. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 59.3% to 70.3%. Qwen3.6-27B does hit back in coding, so the answer changes if that is the part of the workload you care about most.
Qwen3.7 Plus gives you the larger context window at 1M, compared with 262K for Qwen3.6-27B.
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 | Qwen3.6-27B | Δ | Qwen3.7 Plus |
|---|---|---|---|
| Agentic | Qwen3.6-27B59.3 | Margin→ 12.4 | Qwen3.7 Plus71.7 |
| Knowledge | Qwen3.6-27B53.3 | Margin→ 6.8 | Qwen3.7 Plus60.1 |
| Multimodal | Qwen3.6-27B76.7 | Margin→ 4.8 | Qwen3.7 Plus81.5 |
| Math | Qwen3.6-27B89.2 | Margin→ 3.7 | Qwen3.7 Plus92.9 |
| Coding | Qwen3.6-27B77.5 | Margin← 1.9 | Qwen3.7 Plus75.6 |
| Reasoning | Qwen3.6-27BNot measured | MarginNo overlap | Qwen3.7 Plus91.7 |
| Multilingual | Qwen3.6-27BNot measured | MarginNo overlap | Qwen3.7 Plus85.4 |
| Inst. Following | Qwen3.6-27BNot 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 59.3%B 70.3%Winner: Qwen3.7 PlusΔ 11Terminal-Bench 2.0: Qwen3.6-27B scored 59.3%; Qwen3.7 Plus scored 70.3%. Qwen3.7 Plus wins this benchmark. - Source ↗
HLE
KnowledgeA 24%B 34.7%Winner: Qwen3.7 PlusΔ 10.7HLE: Qwen3.6-27B scored 24%; Qwen3.7 Plus scored 34.7%. Qwen3.7 Plus wins this benchmark. - Source ↗
HMMT Feb 2026
MathA 84.3%B 92.9%Winner: Qwen3.7 PlusΔ 8.6HMMT Feb 2026: Qwen3.6-27B scored 84.3%; Qwen3.7 Plus scored 92.9%. Qwen3.7 Plus wins this benchmark. - Source ↗
CharXiv
MultimodalA 78.4%B 85.9%Winner: Qwen3.7 PlusΔ 7.5CharXiv: Qwen3.6-27B scored 78.4%; Qwen3.7 Plus scored 85.9%. Qwen3.7 Plus wins this benchmark. - Source ↗
LiveCodeBench
CodingA 83.9%B 89.6%Winner: Qwen3.7 PlusΔ 5.7LiveCodeBench: Qwen3.6-27B scored 83.9%; Qwen3.7 Plus scored 89.6%. Qwen3.7 Plus wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Qwen3.6-27B | Qwen3.7 Plus | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Qwen3.6-27B$0 input / $0 output | Qwen3.7 PlusNot available | A complete price comparison is not available. |
| Generation speedtokens per second | Qwen3.6-27BNot available | Qwen3.7 PlusNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Qwen3.6-27BNot available | Qwen3.7 PlusNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Qwen3.6-27B262K | Qwen3.7 Plus1M | Qwen3.7 Plus lists the larger context window. |
Benchmark Deep Dive
AgenticQwen3.7 Plus wins17 benchmarks
| Benchmark | Qwen3.6-27B | Qwen3.7 Plus | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 59.3% | 70.3% | Qwen3.7 Plus leads |
| Claw-EvalSource | 72.4% | 62.7% | Qwen3.6-27B leads |
| QwenClawBenchSource | 53.4% | 61.8% | Qwen3.7 Plus leads |
| QwenWebBenchSource | 1487 | 1536 | Qwen3.7 Plus leads |
| AndroidWorldSource | 70.3% | 81.0% | Qwen3.7 Plus leads |
| AA Agentic IndexSource | 27.0% | 20.8% | Qwen3.6-27B leads |
| τ²-bench resultsSource | 94.2% | 93% | Qwen3.6-27B leads |
| GDPval-AASource | 32.0% | 21.8% | Qwen3.6-27B leads |
| GDPval-AASource | 1140 | 936 | Qwen3.6-27B leads |
| Gert LabsSource | 54.84% | — | Not comparable |
| BFCL v4Source | — | 72.9% | Not comparable |
| MCP AtlasSource | — | 73.2% | Not comparable |
| VITA-BenchSource | — | 45.6% | Not comparable |
| DeepPlanningSource | — | 62.3% | Not comparable |
| OSWorld-VerifiedSource | — | 73.3% | Not comparable |
| APEX-Agents-AASource | — | 22.4% | Not comparable |
| OSWorld 2.0Source | — | 2.8% | Not comparable |
CodingQwen3.6-27B wins9 benchmarks
| Benchmark | Qwen3.6-27B | Qwen3.7 Plus | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 77.2% | 77.7% | Qwen3.7 Plus leads |
| SWE MultilingualSource | 71.3% | 75.8% | Qwen3.7 Plus leads |
| SWE-bench ProSource | 53.5% | 57.6% | Qwen3.7 Plus leads |
| Terminal-Bench 2.0Source | 59.3% | 70.3% | Qwen3.7 Plus leads |
| LiveCodeBenchSource | 83.9% | 89.6% | Qwen3.7 Plus leads |
| NL2RepoSource | 36.2% | 41.1% | Qwen3.7 Plus leads |
| AA Coding IndexSource | 53.7% | 55.9% | Qwen3.7 Plus leads |
| AA-SciCodeSource | 39.8% | 45.5% | Qwen3.7 Plus leads |
| SciCodeSource | — | 51.3% | Not comparable |
Reasoning3 benchmarks
KnowledgeQwen3.7 Plus wins14 benchmarks
| Benchmark | Qwen3.6-27B | Qwen3.7 Plus | Result |
|---|---|---|---|
| MMLU-ProSource | 86.2% | 88.5% | Qwen3.7 Plus leads |
| MMLU-ReduxSource | 93.5% | 94.5% | Qwen3.7 Plus leads |
| SuperGPQASource | 66% | 71.4% | Qwen3.7 Plus leads |
| C-EvalSource | 91.4% | — | Not comparable |
| GPQASource | 87.8% | 90.3% | Qwen3.7 Plus leads |
| HLESource | 24% | 34.7% | Qwen3.7 Plus leads |
| Artificial Analysis Intelligence IndexSource | 37.0% | 39.0% | Qwen3.7 Plus leads |
| AA-GPQA DiamondSource | 84.2% | 90.0% | Qwen3.7 Plus leads |
| AA-HLESource | 21.6% | 33.4% | Qwen3.7 Plus leads |
| AA-Omniscience IndexSource | -19.8% | 2.4% | Qwen3.7 Plus leads |
| AA-Omniscience AccuracySource | 19.2% | 22.2% | Qwen3.7 Plus leads |
| AA-Omniscience Hallucination RateSource | 48.3% | 25.5% | Qwen3.7 Plus leads |
| GPQA-DSource | — | 90.3% | Not comparable |
| MMMLUSource | — | 89.0% | Not comparable |
MathQwen3.7 Plus wins7 benchmarks
| Benchmark | Qwen3.6-27B | Qwen3.7 Plus | Result |
|---|---|---|---|
| HMMT Feb 2025Source | 93.8% | — | Not comparable |
| HMMT Nov 2025Source | 90.7% | — | Not comparable |
| HMMT Feb 2026Source | 84.3% | 92.9% | Qwen3.7 Plus leads |
| MMAnswerBenchSource | 80.8% | — | Not comparable |
| AIME26Source | 94.1% | — | Not comparable |
| IMOAnswerBenchSource | — | 86.0% | Not comparable |
| ApexSource | — | 22.7% | Not comparable |
Multilingual5 benchmarks
MultimodalQwen3.7 Plus wins24 benchmarks
| Benchmark | Qwen3.6-27B | Qwen3.7 Plus | Result |
|---|---|---|---|
| MMMUSource | 82.9% | — | Not comparable |
| MMMU-ProSource | 75.8% | 79% | Qwen3.7 Plus leads |
| RealWorldQASource | 84.1% | 86.9% | Qwen3.7 Plus leads |
| DynaMathSource | 85.6% | — | Not comparable |
| MStarSource | 81.4% | — | Not comparable |
| SimpleVQASource | 56.1% | 81.7% | Qwen3.7 Plus leads |
| CharXivSource | 78.4% | 85.9% | Qwen3.7 Plus leads |
| CC-OCRSource | 81.2% | — | Not comparable |
| CountBenchSource | 97.8% | — | Not comparable |
| RefCOCO (avg)Source | 92.5% | — | Not comparable |
| ERQASource | 62.5% | 69.8% | Qwen3.7 Plus leads |
| Video-MME (with subtitle)Source | 87.7% | 88.0% | Qwen3.7 Plus leads |
| VideoMMMUSource | 84.4% | 85.4% | Qwen3.7 Plus leads |
| MLVU (M-Avg)Source | 86.6% | 87.4% | Qwen3.7 Plus leads |
| V*Source | 94.7% | — | Not comparable |
| AA-MMMU-ProSource | 74.6% | 80.5% | Qwen3.7 Plus leads |
| MathVisionSource | — | 90.3% | Not comparable |
| MedXpertQA (MM)Source | — | 71.0% | Not comparable |
| ScreenSpot ProSource | — | 79.0% | Not comparable |
| MMSearch-PlusSource | — | 41.4% | Not comparable |
| OmniDocBench 1.5Source | — | 91.4% | Not comparable |
| OCRBench V2Source | — | 70.7% | Not comparable |
| ODINW13Source | — | 51.1% | Not comparable |
| Design Arena WebsiteSource | — | 1288 | Not comparable |
Frequently Asked Questions (6)
Which is better, Qwen3.6-27B or Qwen3.7 Plus?
Qwen3.7 Plus is ahead on BenchLM's BenchAlign leaderboard, 67.22 to 53.82. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 59.3% and 70.3%.
Which is better for knowledge tasks, Qwen3.6-27B or Qwen3.7 Plus?
Qwen3.7 Plus has the edge for knowledge tasks in this comparison, averaging 60.1 versus 53.3. Inside this category, AA-Omniscience Hallucination Rate is the benchmark that creates the most daylight between them.
Which is better for coding, Qwen3.6-27B or Qwen3.7 Plus?
Qwen3.6-27B has the edge for coding in this comparison, averaging 77.5 versus 75.6. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.
Which is better for math, Qwen3.6-27B or Qwen3.7 Plus?
Qwen3.7 Plus has the edge for math in this comparison, averaging 92.9 versus 89.2. Inside this category, HMMT Feb 2026 is the benchmark that creates the most daylight between them.
Which is better for agentic tasks, Qwen3.6-27B or Qwen3.7 Plus?
Qwen3.7 Plus has the edge for agentic tasks in this comparison, averaging 71.7 versus 59.3. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.
Which is better for multimodal and grounded tasks, Qwen3.6-27B or Qwen3.7 Plus?
Qwen3.7 Plus has the edge for multimodal and grounded tasks in this comparison, averaging 81.5 versus 76.7. Inside this category, SimpleVQA is the benchmark that creates the most daylight between them.
Self-host vs API cost
Estimates at 50,000 req/day · 1000 tokens/req average.
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.