# DeepSeek V4.1 Flash vs Qwen3.5-Flash

> DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 42.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v4-1-flash-vs-qwen3-5-flash
- Last updated: 2026-10-12
- Shared benchmarks: 26

## Summary

- They share 26 benchmarks with published results for both. DeepSeek V4.1 Flash 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 V4.1 Flash leads 66.7 to 37.4.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 67.4% for DeepSeek V4.1 Flash and 18.2% for Qwen3.5-Flash.
- Qwen3.5-Flash is cheaper at $0.10 / $0.40 per million input/output tokens, against $0.15 / $0.60 for DeepSeek V4.1 Flash.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek V4.1 Flash | Qwen3.5-Flash |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 52.8 | 42.5 |
| Rank | 38 | 112 |
| Context | 1M | 1M |
| Input $/M | $0.15 | $0.10 |
| Output $/M | $0.60 | $0.40 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek V4.1 Flash: 52.9 (#32)
- Qwen3.5-Flash: 34.2 (#242)

| Benchmark | DeepSeek V4.1 Flash | Qwen3.5-Flash |
|---|---|---|
| LMArena WebDev | 1619 | 1244 |
| LMArena Coding | 1506 | 1412 |
| ALE-Bench | 1,092 | 221.8 |
| SciCode | 51.9% | — |

## Agentic & Tool Use

- DeepSeek V4.1 Flash: 31.2 (#69)
- Qwen3.5-Flash: —

| Benchmark | DeepSeek V4.1 Flash | Qwen3.5-Flash |
|---|---|---|
| APEX-Agents | 39.5% | — |
| GDP.pdf | 19.8% | — |
| Vending-Bench 2 | — | 462.69 |

## Reasoning

- DeepSeek V4.1 Flash: 50.2 (#36)
- Qwen3.5-Flash: 33.7 (#72)

| Benchmark | DeepSeek V4.1 Flash | Qwen3.5-Flash |
|---|---|---|
| LMArena Hard Prompts | 1483 | 1403 |
| Mystery Game Puzzles | 43% | 20% |
| DTBench | 89.9% | 82.9% |
| LMCA | 47% | 29.1% |
| Epoch Capabilities Index | 154.9 | 143.98 |
| NYT Connections (extended) | 89.6% | — |
| CritPt | 14.3% | — |
| Chess Puzzles | — | 21% |
| Surface Evolver Bench | 46.3% | — |

## Math

- DeepSeek V4.1 Flash: 66.7 (#25)
- Qwen3.5-Flash: 37.4 (#158)

| Benchmark | DeepSeek V4.1 Flash | Qwen3.5-Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 67.4% | 18.2% |
| OTIS Mock AIME 2024-2025 | 98.3% | 84.4% |
| LMArena Math | 1477 | 1407 |
| FrontierMath Tier 4 | 26.8% | — |
| ProofBench | 54% | — |
| FrontierMath (Feb 2025 set) | — | 6.2% |
| FrontierMath Tier 4 (v1) | — | 0% |

## Knowledge

- DeepSeek V4.1 Flash: 57.9 (#38)
- Qwen3.5-Flash: 43.2 (#93)

| Benchmark | DeepSeek V4.1 Flash | Qwen3.5-Flash |
|---|---|---|
| GPQA Diamond | 89.8% | 82.3% |
| LMArena Expert | 1506 | 1407 |
| SimpleQA Verified | — | 20.3% |
| Vectara Hallucination Rate | — | 10.5% |

## Multimodal

- DeepSeek V4.1 Flash: 39.1 (#61)
- Qwen3.5-Flash: —

| Benchmark | DeepSeek V4.1 Flash | Qwen3.5-Flash |
|---|---|---|
| LMArena Vision | 1277 | — |
| Furniture Assembly | 34.2% | — |

## Multilingual

- DeepSeek V4.1 Flash: 55.0 (#35)
- Qwen3.5-Flash: 50.5 (#121)

| Benchmark | DeepSeek V4.1 Flash | Qwen3.5-Flash |
|---|---|---|
| LMArena Non-English | 1448 | 1385 |
| LMArena Chinese | 1497 | 1446 |
| LMArena French | 1452 | 1412 |
| LMArena German | 1484 | 1390 |
| LMArena Japanese | 1412 | 1368 |
| LMArena Korean | 1452 | 1344 |
| LMArena Russian | 1471 | 1379 |
| LMArena Spanish | 1459 | 1400 |

## Instruction Following

- DeepSeek V4.1 Flash: 77.3 (#26)
- Qwen3.5-Flash: 72.6 (#139)

| Benchmark | DeepSeek V4.1 Flash | Qwen3.5-Flash |
|---|---|---|
| LMArena Instruction Following | 1474 | 1374 |

## Long Context

- DeepSeek V4.1 Flash: 45.2 (#47)
- Qwen3.5-Flash: 42.4 (#124)

| Benchmark | DeepSeek V4.1 Flash | Qwen3.5-Flash |
|---|---|---|
| LMArena Longer Query | 1475 | 1392 |

## Writing & Preference

- DeepSeek V4.1 Flash: 65.4 (#48)
- Qwen3.5-Flash: 57.9 (#122)

| Benchmark | DeepSeek V4.1 Flash | Qwen3.5-Flash |
|---|---|---|
| LMArena Text | 1462 | 1397 |
| LMArena Creative Writing | 1435 | 1343 |
| LMArena Multi-Turn | 1457 | 1393 |
| EQ-Bench Creative Writing | 1540 | — |

## FAQ

### Is DeepSeek V4.1 Flash better than Qwen3.5-Flash?

DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 42.5 on the Noometry Index.

### Which is cheaper, DeepSeek V4.1 Flash 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 V4.1 Flash lists at $0.15 and $0.60.

### Is DeepSeek V4.1 Flash or Qwen3.5-Flash better for coding?

DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 34.2 in the Noometry coding category.

### Which has the bigger context window?

Both accept 1M tokens.

### How many benchmarks do DeepSeek V4.1 Flash and Qwen3.5-Flash share?

26 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and Qwen3.5-Flash has 32.