Model comparison
Qwen3.6 Plus vs Qwen3.8 27B
Qwen3.6 Plus is the stronger model overall, scoring 47.5 to 46.0 on the Noometry Index.
Last verified . 24 shared benchmarks.
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. Qwen3.6 Plus scores higher in 3 categories and Qwen3.8 27B in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.6 Plus leads 51.8 to 37.1.
- The biggest single-benchmark swing is LMCA: 33.1% for Qwen3.6 Plus and 41.4% for Qwen3.8 27B.
- Both cost about the same: $0.50 input and $3 output per million tokens.
- Qwen3.6 Plus accepts more context: 1M tokens versus 262K.
- Qwen3.8 27B has downloadable open weights; the other is API-only.
Side by side
| Qwen3.6 Plus | Qwen3.8 27B | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 47.5 | 46.0 |
| Released | 2026-03-31 | 2026-08-14 |
| Weights | Proprietary | Open |
| Context window | 1M | 262K |
| Max output | 66K | 33K |
| Input $ / M tokens | $0.50 | $0.99 |
| Output $ / M tokens | $3 | $1.49 |
| Results tracked | 37 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
Qwen3.6 Plus: 40.8 (#130), Qwen3.8 27B: 50.5 (#44)
| Benchmark | Qwen3.6 Plus | Qwen3.8 27B |
|---|---|---|
| LMArena WebDev | 1461 | 1593 |
| SciCode | 40.7% | 46.6% |
| LMArena Coding | 1467 | 1482 |
| SWE-bench Verified | 57.9% | — |
| ALE-Bench | 670.15 | — |
Agentic & Tool Use Not comparable
Qwen3.6 Plus: —, Qwen3.8 27B: 32.9 (#57)
| Benchmark | Qwen3.6 Plus | Qwen3.8 27B |
|---|---|---|
| APEX-Agents | — | 47.5% |
| Vending-Bench 2 | 5,115 | — |
Reasoning Qwen3.8 27B leads
Qwen3.6 Plus: 29.3 (#93), Qwen3.8 27B: 41.0 (#54)
| Benchmark | Qwen3.6 Plus | Qwen3.8 27B |
|---|---|---|
| NYT Connections (extended) | 60.3% | 54.5% |
| CritPt | 2.9% | 5.4% |
| LMArena Hard Prompts | 1449 | 1460 |
| DTBench | 81.9% | 88% |
| LMCA | 33.1% | 41.4% |
| Epoch Capabilities Index | 147.65 | 149.38 |
| ARC-AGI-2 | — | 42.4% |
| ARC-AGI-1 | — | 87.5% |
| Chess Puzzles | 17% | — |
| Thematic Generalization | 59.5% | — |
| Mystery Game Puzzles | 12% | — |
| Surface Evolver Bench | — | 45% |
Math Qwen3.6 Plus leads
Qwen3.6 Plus: 51.8 (#54), Qwen3.8 27B: 37.1 (#161)
| Benchmark | Qwen3.6 Plus | Qwen3.8 27B |
|---|---|---|
| LMArena Math | 1450 | 1456 |
| FrontierMath (Tiers 1-3) | 38.2% | — |
| OTIS Mock AIME 2024-2025 | 93.3% | — |
| ProofBench | — | 16% |
| FrontierMath (Feb 2025 set) | 26.2% | — |
| FrontierMath Tier 4 (v1) | 8.3% | — |
Knowledge Qwen3.6 Plus leads
Qwen3.6 Plus: 56.1 (#45), Qwen3.8 27B: 41.6 (#109)
| Benchmark | Qwen3.6 Plus | Qwen3.8 27B |
|---|---|---|
| LMArena Expert | 1454 | 1482 |
| GPQA Diamond | 88.4% | — |
| SimpleQA Verified | 44.1% | — |
Multimodal Not comparable
Qwen3.6 Plus: —, Qwen3.8 27B: 41.3 (#37)
| Benchmark | Qwen3.6 Plus | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | — | 1271 |
Multilingual Too close to call
Qwen3.6 Plus: 53.3 (#70), Qwen3.8 27B: 53.7 (#60)
| Benchmark | Qwen3.6 Plus | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | 1424 | 1430 |
| LMArena Chinese | 1477 | 1504 |
| LMArena French | 1455 | 1465 |
| LMArena German | 1452 | 1438 |
| LMArena Japanese | 1389 | 1384 |
| LMArena Korean | 1379 | 1393 |
| LMArena Russian | 1434 | 1415 |
| LMArena Spanish | 1432 | 1448 |
Instruction Following Too close to call
Qwen3.6 Plus: 75.0 (#74), Qwen3.8 27B: 75.8 (#53)
| Benchmark | Qwen3.6 Plus | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | 1425 | 1439 |
Long Context Too close to call
Qwen3.6 Plus: 45.2 (#49), Qwen3.8 27B: 44.3 (#70)
| Benchmark | Qwen3.6 Plus | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | 1439 | 1450 |
| CL-bench | 20.3% | — |
Writing & Preference Qwen3.8 27B leads
Qwen3.6 Plus: 62.2 (#82), Qwen3.8 27B: 65.8 (#43)
| Benchmark | Qwen3.6 Plus | Qwen3.8 27B |
|---|---|---|
| LMArena Text | 1437 | 1441 |
| LMArena Creative Writing | 1404 | 1384 |
| LMArena Multi-Turn | 1438 | 1441 |
| EQ-Bench Creative Writing | — | 1671 |
Frequently asked questions
Is Qwen3.6 Plus better than Qwen3.8 27B?
Qwen3.6 Plus is the stronger model overall, scoring 47.5 to 46.0 on the Noometry Index.
Which is cheaper, Qwen3.6 Plus or Qwen3.8 27B?
Qwen3.8 27B is cheaper. It lists at $0.99 per million input tokens and $1.49 per million output tokens; Qwen3.6 Plus lists at $0.50 and $3.
Is Qwen3.6 Plus or Qwen3.8 27B better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.5 versus 40.8 in the Noometry coding category.
Which has the bigger context window?
Qwen3.6 Plus does, with 1M tokens against 262K.
How many benchmarks do Qwen3.6 Plus and Qwen3.8 27B share?
24 benchmarks have published results for both models. Qwen3.6 Plus has 37 scored results on Noometry and Qwen3.8 27B has 31.