Model comparison
DeepSeek-V3 vs Qwen3.6 Plus
Qwen3.6 Plus is the stronger model overall, scoring 47.5 to 39.5 on the Noometry Index. DeepSeek-V3 costs 2.8× less per token, which makes it the better buy when Qwen3.6 Plus's lead doesn't matter for your workload.
Last verified . 25 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 25 benchmarks with published results for both. DeepSeek-V3 scores higher in 1 category and Qwen3.6 Plus in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.6 Plus leads 51.8 to 32.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for DeepSeek-V3 and 93.3% for Qwen3.6 Plus.
- DeepSeek-V3 is cheaper at $0.24 / $0.90 per million input/output tokens, against $0.50 / $3 for Qwen3.6 Plus.
- Qwen3.6 Plus accepts more context: 1M tokens versus 164K.
- DeepSeek-V3 has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3 | Qwen3.6 Plus | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 39.5 | 47.5 |
| Released | 2024-12-26 | 2026-03-31 |
| Weights | Open | Proprietary |
| Context window | 164K | 1M |
| Max output | 164K | 66K |
| Input $ / M tokens | $0.24 | $0.50 |
| Output $ / M tokens | $0.90 | $3 |
| Results tracked | 60 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3 leads
DeepSeek-V3: 42.3 (#106), Qwen3.6 Plus: 40.8 (#130)
| Benchmark | DeepSeek-V3 | Qwen3.6 Plus |
|---|---|---|
| SciCode | 35.8% | 40.7% |
| LMArena Coding | 1368 | 1467 |
| SWE-bench Verified | — | 57.9% |
| Aider Polyglot | 55.1% | — |
| LMArena WebDev | — | 1461 |
| WeirdML | 36.1% | — |
| BigCodeBench Instruct | 50% | — |
| LiveBench Coding | 70.9% | — |
| BigCodeBench Complete | 62.2% | — |
| ALE-Bench | — | 670.15 |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73% | — |
Agentic & Tool Use Not comparable
DeepSeek-V3: —, Qwen3.6 Plus: —
| Benchmark | DeepSeek-V3 | Qwen3.6 Plus |
|---|---|---|
| METR Time Horizons | 49.6% | — |
| Vending-Bench 2 | — | 5,115 |
Reasoning Qwen3.6 Plus leads
DeepSeek-V3: 20.5 (#236), Qwen3.6 Plus: 29.3 (#93)
| Benchmark | DeepSeek-V3 | Qwen3.6 Plus |
|---|---|---|
| CritPt | 0% | 2.9% |
| LMArena Hard Prompts | 1365 | 1449 |
| DTBench | 64.8% | 81.9% |
| LMCA | 15.5% | 33.1% |
| Epoch Capabilities Index | 135.94 | 147.65 |
| SimpleBench | 27.2% | — |
| Kagi LLM Benchmark | 52.3% | — |
| NYT Connections (extended) | — | 60.3% |
| Chess Puzzles | — | 17% |
| Thematic Generalization | — | 59.5% |
| LiveBench Reasoning | 65.8% | — |
| Mystery Game Puzzles | — | 12% |
| LiveBench Data Analysis | 60.9% | — |
| BIG-Bench Hard | 87.5% | — |
| ForecastBench | 59.1 | — |
| HellaSwag | 88.9% | — |
| LiveBench | 66.9% | — |
| PIQA | 84.7% | — |
| WinoGrande | 85.2% | — |
Math Qwen3.6 Plus leads
DeepSeek-V3: 32.1 (#219), Qwen3.6 Plus: 51.8 (#54)
| Benchmark | DeepSeek-V3 | Qwen3.6 Plus |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 93.3% |
| LMArena Math | 1373 | 1450 |
| FrontierMath (Feb 2025 set) | 1.7% | 26.2% |
| FrontierMath (Tiers 1-3) | — | 38.2% |
| Omni-MATH | 40.3% | — |
| LiveBench Math | 73.5% | — |
| MATH Level 5 | 75.5% | — |
| FrontierMath Tier 4 (v1) | — | 8.3% |
Knowledge Qwen3.6 Plus leads
DeepSeek-V3: 37.5 (#155), Qwen3.6 Plus: 56.1 (#45)
| Benchmark | DeepSeek-V3 | Qwen3.6 Plus |
|---|---|---|
| GPQA Diamond | 67.6% | 88.4% |
| LMArena Expert | 1351 | 1454 |
| SimpleQA Verified | — | 44.1% |
| MMLU-Pro | 72.3% | — |
| Confabulations | 26.1% | — |
| Vectara Hallucination Rate | 6.1% | — |
| GPQA (HELM) | 53.8% | — |
| ARC (AI2) Challenge | 95.3% | — |
| MMLU | 87.2% | — |
| TriviaQA | 82.9% | — |
Multilingual Qwen3.6 Plus leads
DeepSeek-V3: 48.5 (#143), Qwen3.6 Plus: 53.3 (#70)
| Benchmark | DeepSeek-V3 | Qwen3.6 Plus |
|---|---|---|
| LMArena Non-English | 1358 | 1424 |
| LMArena Chinese | 1391 | 1477 |
| LMArena French | 1385 | 1455 |
| LMArena German | 1374 | 1452 |
| LMArena Japanese | 1333 | 1389 |
| LMArena Korean | 1319 | 1379 |
| LMArena Russian | 1373 | 1434 |
| LMArena Spanish | 1358 | 1432 |
Instruction Following Qwen3.6 Plus leads
DeepSeek-V3: 72.8 (#130), Qwen3.6 Plus: 75.0 (#74)
| Benchmark | DeepSeek-V3 | Qwen3.6 Plus |
|---|---|---|
| LMArena Instruction Following | 1345 | 1425 |
| LiveBench Instruction Following | 81.5% | — |
| IFEval | 83.2% | — |
Long Context Qwen3.6 Plus leads
DeepSeek-V3: 34.0 (#253), Qwen3.6 Plus: 45.2 (#49)
| Benchmark | DeepSeek-V3 | Qwen3.6 Plus |
|---|---|---|
| LMArena Longer Query | 1352 | 1439 |
| Fiction.LiveBench | 50% | — |
| CL-bench | — | 20.3% |
Writing & Preference Qwen3.6 Plus leads
DeepSeek-V3: 57.4 (#130), Qwen3.6 Plus: 62.2 (#82)
| Benchmark | DeepSeek-V3 | Qwen3.6 Plus |
|---|---|---|
| LMArena Text | 1375 | 1437 |
| LMArena Creative Writing | 1364 | 1404 |
| LMArena Multi-Turn | 1389 | 1438 |
| Short-Story Creative Writing | 77% | — |
| EQ-Bench Creative Writing | 1472 | — |
| WildBench | 83% | — |
| LiveBench Language | 49.1% | — |
Frequently asked questions
Is DeepSeek-V3 better than Qwen3.6 Plus?
Qwen3.6 Plus is the stronger model overall, scoring 47.5 to 39.5 on the Noometry Index. DeepSeek-V3 costs 2.8× less per token, which makes it the better buy when Qwen3.6 Plus's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3 or Qwen3.6 Plus?
DeepSeek-V3 is cheaper. It lists at $0.24 per million input tokens and $0.90 per million output tokens; Qwen3.6 Plus lists at $0.50 and $3.
Is DeepSeek-V3 or Qwen3.6 Plus better for coding?
DeepSeek-V3 scores higher on coding benchmarks: 42.3 versus 40.8 in the Noometry coding category.
Which has the bigger context window?
Qwen3.6 Plus does, with 1M tokens against 164K.
How many benchmarks do DeepSeek-V3 and Qwen3.6 Plus share?
25 benchmarks have published results for both models. DeepSeek-V3 has 60 scored results on Noometry and Qwen3.6 Plus has 37.