Model comparison
DeepSeek V4 Pro vs Qwen Plus
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 37.1 on the Noometry Index. Qwen Plus costs 1.6× less per token, which makes it the better buy when DeepSeek V4 Pro's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 18 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 8 categories and Qwen Plus in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4 Pro leads 64.8 to 23.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.6% for DeepSeek V4 Pro and 17.8% for Qwen Plus.
- Qwen Plus is cheaper at $0.40 / $1.20 per million input/output tokens, against $0.66 / $1.98 for DeepSeek V4 Pro.
- DeepSeek V4 Pro has downloadable open weights; the other is API-only.
Side by side
| DeepSeek V4 Pro | Qwen Plus | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 54.3 | 37.1 |
| Released | 2026-04-24 | 2024-01-25 |
| Weights | Open | Proprietary |
| Context window | 1M | 1M |
| Max output | 393K | 33K |
| Input $ / M tokens | $0.66 | $0.40 |
| Output $ / M tokens | $1.98 | $1.20 |
| Results tracked | 48 | 20 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Pro leads
DeepSeek V4 Pro: 52.4 (#34), Qwen Plus: 38.9 (#167)
| Benchmark | DeepSeek V4 Pro | Qwen Plus |
|---|---|---|
| LMArena Coding | 1470 | 1328 |
| SWE-bench Verified | 77.6% | — |
| FrontierCode | 28.6% | — |
| LMArena WebDev | 1582 | — |
| SciCode | 51% | — |
| WeirdML | 66.2% | — |
| ALE-Bench | 1,403 | — |
Agentic & Tool Use Not comparable
DeepSeek V4 Pro: 32.8 (#58), Qwen Plus: —
| Benchmark | DeepSeek V4 Pro | Qwen Plus |
|---|---|---|
| APEX-Agents | 47.3% | — |
| Vending-Bench 2 | 3,285 | — |
Reasoning DeepSeek V4 Pro leads
DeepSeek V4 Pro: 56.5 (#24), Qwen Plus: 28.4 (#107)
| Benchmark | DeepSeek V4 Pro | Qwen Plus |
|---|---|---|
| Kagi LLM Benchmark | 53.5% | 63.3% |
| LMArena Hard Prompts | 1461 | 1317 |
| DTBench | 93.9% | 81.1% |
| LMCA | 45.5% | 24% |
| ARC-AGI-2 | 61.3% | — |
| NYT Connections (extended) | 91.3% | — |
| ARC-AGI-1 | 90.5% | — |
| CritPt | 18% | — |
| Chess Puzzles | 47% | — |
| Mystery Game Puzzles | 43% | — |
| Surface Evolver Bench | 40% | — |
| Epoch Capabilities Index | 155.31 | — |
| ForecastBench | 56.1 | — |
Math DeepSeek V4 Pro leads
DeepSeek V4 Pro: 64.8 (#30), Qwen Plus: 23.3 (#271)
| Benchmark | DeepSeek V4 Pro | Qwen Plus |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.6% | 17.8% |
| LMArena Math | 1455 | 1326 |
| FrontierMath (Tiers 1-3) | 64.6% | — |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 76.6% | — |
| ProofBench | 50% | — |
| MATH Level 5 | — | 65.3% |
| FrontierMath (Feb 2025 set) | — | 1.7% |
Knowledge DeepSeek V4 Pro leads
DeepSeek V4 Pro: 59.5 (#31), Qwen Plus: 27.4 (#251)
| Benchmark | DeepSeek V4 Pro | Qwen Plus |
|---|---|---|
| GPQA Diamond | 91.7% | 48.1% |
| LMArena Expert | 1464 | 1328 |
| SimpleQA Verified | 52.9% | — |
| Vectara Hallucination Rate | 8.6% | — |
Multilingual DeepSeek V4 Pro leads
DeepSeek V4 Pro: 54.4 (#45), Qwen Plus: 45.1 (#175)
| Benchmark | DeepSeek V4 Pro | Qwen Plus |
|---|---|---|
| LMArena Non-English | 1439 | 1310 |
| LMArena Chinese | 1486 | 1347 |
| LMArena Japanese | 1445 | 1251 |
| LMArena Russian | 1453 | 1323 |
| LMArena French | 1472 | — |
| LMArena German | 1458 | — |
| LMArena Korean | 1447 | — |
| LMArena Spanish | 1458 | — |
Instruction Following DeepSeek V4 Pro leads
DeepSeek V4 Pro: 76.1 (#47), Qwen Plus: 68.8 (#181)
| Benchmark | DeepSeek V4 Pro | Qwen Plus |
|---|---|---|
| LMArena Instruction Following | 1448 | 1303 |
Long Context DeepSeek V4 Pro leads
DeepSeek V4 Pro: 45.0 (#51), Qwen Plus: 40.3 (#158)
| Benchmark | DeepSeek V4 Pro | Qwen Plus |
|---|---|---|
| LMArena Longer Query | 1458 | 1324 |
| CL-bench Life | 13.5% | — |
Writing & Preference DeepSeek V4 Pro leads
DeepSeek V4 Pro: 65.5 (#46), Qwen Plus: 52.2 (#176)
| Benchmark | DeepSeek V4 Pro | Qwen Plus |
|---|---|---|
| LMArena Text | 1451 | 1326 |
| LMArena Creative Writing | 1446 | 1293 |
| LMArena Multi-Turn | 1467 | 1336 |
| EQ-Bench Creative Writing | 1553 | — |
| EQ-Bench 4 | 1166 | — |
Frequently asked questions
Is DeepSeek V4 Pro better than Qwen Plus?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 37.1 on the Noometry Index. Qwen Plus costs 1.6× less per token, which makes it the better buy when DeepSeek V4 Pro's lead doesn't matter for your workload.
Which is cheaper, DeepSeek V4 Pro or Qwen Plus?
Qwen Plus is cheaper. It lists at $0.40 per million input tokens and $1.20 per million output tokens; DeepSeek V4 Pro lists at $0.66 and $1.98.
Is DeepSeek V4 Pro or Qwen Plus better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 38.9 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do DeepSeek V4 Pro and Qwen Plus share?
18 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Qwen Plus has 20.