Model comparison
DeepSeek V4 Pro vs Qwen3.5-Flash
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 42.5 on the Noometry Index. Qwen3.5-Flash costs 5.7× less per token, which makes it the better buy when DeepSeek V4 Pro's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 30 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 8 categories and Qwen3.5-Flash in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4 Pro leads 64.8 to 37.4.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 64.6% for DeepSeek V4 Pro and 18.2% for Qwen3.5-Flash.
- Qwen3.5-Flash is cheaper at $0.10 / $0.40 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 | Qwen3.5-Flash | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 54.3 | 42.5 |
| Released | 2026-04-24 | 2026-02-23 |
| Weights | Open | Proprietary |
| Context window | 1M | 1M |
| Max output | 393K | 66K |
| Input $ / M tokens | $0.66 | $0.10 |
| Output $ / M tokens | $1.98 | $0.40 |
| Results tracked | 48 | 32 |
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), Qwen3.5-Flash: 34.2 (#242)
| Benchmark | DeepSeek V4 Pro | Qwen3.5-Flash |
|---|---|---|
| LMArena WebDev | 1582 | 1244 |
| LMArena Coding | 1470 | 1412 |
| ALE-Bench | 1,403 | 221.8 |
| SWE-bench Verified | 77.6% | — |
| FrontierCode | 28.6% | — |
| SciCode | 51% | — |
| WeirdML | 66.2% | — |
Agentic & Tool Use Not comparable
DeepSeek V4 Pro: 32.8 (#58), Qwen3.5-Flash: —
| Benchmark | DeepSeek V4 Pro | Qwen3.5-Flash |
|---|---|---|
| Vending-Bench 2 | 3,285 | 462.69 |
| APEX-Agents | 47.3% | — |
Reasoning DeepSeek V4 Pro leads
DeepSeek V4 Pro: 56.5 (#24), Qwen3.5-Flash: 33.7 (#72)
| Benchmark | DeepSeek V4 Pro | Qwen3.5-Flash |
|---|---|---|
| Chess Puzzles | 47% | 21% |
| LMArena Hard Prompts | 1461 | 1403 |
| Mystery Game Puzzles | 43% | 20% |
| DTBench | 93.9% | 82.9% |
| LMCA | 45.5% | 29.1% |
| Epoch Capabilities Index | 155.31 | 143.98 |
| ARC-AGI-2 | 61.3% | — |
| Kagi LLM Benchmark | 53.5% | — |
| NYT Connections (extended) | 91.3% | — |
| ARC-AGI-1 | 90.5% | — |
| CritPt | 18% | — |
| Surface Evolver Bench | 40% | — |
| ForecastBench | 56.1 | — |
Math DeepSeek V4 Pro leads
DeepSeek V4 Pro: 64.8 (#30), Qwen3.5-Flash: 37.4 (#158)
| Benchmark | DeepSeek V4 Pro | Qwen3.5-Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 64.6% | 18.2% |
| OTIS Mock AIME 2024-2025 | 98.6% | 84.4% |
| LMArena Math | 1455 | 1407 |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 76.6% | — |
| ProofBench | 50% | — |
| FrontierMath (Feb 2025 set) | — | 6.2% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge DeepSeek V4 Pro leads
DeepSeek V4 Pro: 59.5 (#31), Qwen3.5-Flash: 43.2 (#93)
| Benchmark | DeepSeek V4 Pro | Qwen3.5-Flash |
|---|---|---|
| GPQA Diamond | 91.7% | 82.3% |
| SimpleQA Verified | 52.9% | 20.3% |
| Vectara Hallucination Rate | 8.6% | 10.5% |
| LMArena Expert | 1464 | 1407 |
Multilingual DeepSeek V4 Pro leads
DeepSeek V4 Pro: 54.4 (#45), Qwen3.5-Flash: 50.5 (#121)
| Benchmark | DeepSeek V4 Pro | Qwen3.5-Flash |
|---|---|---|
| LMArena Non-English | 1439 | 1385 |
| LMArena Chinese | 1486 | 1446 |
| LMArena French | 1472 | 1412 |
| LMArena German | 1458 | 1390 |
| LMArena Japanese | 1445 | 1368 |
| LMArena Korean | 1447 | 1344 |
| LMArena Russian | 1453 | 1379 |
| LMArena Spanish | 1458 | 1400 |
Instruction Following DeepSeek V4 Pro leads
DeepSeek V4 Pro: 76.1 (#47), Qwen3.5-Flash: 72.6 (#139)
| Benchmark | DeepSeek V4 Pro | Qwen3.5-Flash |
|---|---|---|
| LMArena Instruction Following | 1448 | 1374 |
Long Context DeepSeek V4 Pro leads
DeepSeek V4 Pro: 45.0 (#51), Qwen3.5-Flash: 42.4 (#124)
| Benchmark | DeepSeek V4 Pro | Qwen3.5-Flash |
|---|---|---|
| LMArena Longer Query | 1458 | 1392 |
| CL-bench Life | 13.5% | — |
Writing & Preference DeepSeek V4 Pro leads
DeepSeek V4 Pro: 65.5 (#46), Qwen3.5-Flash: 57.9 (#122)
| Benchmark | DeepSeek V4 Pro | Qwen3.5-Flash |
|---|---|---|
| LMArena Text | 1451 | 1397 |
| LMArena Creative Writing | 1446 | 1343 |
| LMArena Multi-Turn | 1467 | 1393 |
| EQ-Bench Creative Writing | 1553 | — |
| EQ-Bench 4 | 1166 | — |
Frequently asked questions
Is DeepSeek V4 Pro better than Qwen3.5-Flash?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 42.5 on the Noometry Index. Qwen3.5-Flash costs 5.7× 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 Qwen3.5-Flash?
Qwen3.5-Flash is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; DeepSeek V4 Pro lists at $0.66 and $1.98.
Is DeepSeek V4 Pro or Qwen3.5-Flash better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 34.2 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do DeepSeek V4 Pro and Qwen3.5-Flash share?
30 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Qwen3.5-Flash has 32.