Model comparison
DeepSeek V4 Flash vs Qwen3 Max
DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 43.7 on the Noometry Index.
Last verified . 29 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 29 benchmarks with published results for both. DeepSeek V4 Flash scores higher in 7 categories and Qwen3 Max in 1 category; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Flash leads 53.7 to 22.6.
- The biggest single-benchmark swing is NYT Connections (extended): 89.6% for DeepSeek V4 Flash and 30.1% for Qwen3 Max.
- DeepSeek V4 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $1.20 / $6 for Qwen3 Max.
- DeepSeek V4 Flash accepts more context: 1M tokens versus 262K.
- DeepSeek V4 Flash has downloadable open weights; the other is API-only.
Side by side
| DeepSeek V4 Flash | Qwen3 Max | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 53.6 | 43.7 |
| Released | 2026-04-24 | 2025-09-23 |
| Weights | Open | Proprietary |
| Context window | 1M | 262K |
| Max output | 393K | 66K |
| Input $ / M tokens | $0.15 | $1.20 |
| Output $ / M tokens | $0.60 | $6 |
| Results tracked | 41 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Flash leads
DeepSeek V4 Flash: 47.9 (#59), Qwen3 Max: 43.0 (#93)
| Benchmark | DeepSeek V4 Flash | Qwen3 Max |
|---|---|---|
| LMArena Coding | 1457 | 1456 |
| ALE-Bench | 1,306 | 370.45 |
| FrontierCode | 18.8% | — |
| LMArena WebDev | 1582 | — |
| SciCode | 49.9% | — |
| WeirdML | 63% | — |
Agentic & Tool Use Not comparable
DeepSeek V4 Flash: —, Qwen3 Max: —
| Benchmark | DeepSeek V4 Flash | Qwen3 Max |
|---|---|---|
| Vending-Bench 2 | — | 71.56 |
Reasoning DeepSeek V4 Flash leads
DeepSeek V4 Flash: 53.7 (#30), Qwen3 Max: 22.6 (#190)
| Benchmark | DeepSeek V4 Flash | Qwen3 Max |
|---|---|---|
| Kagi LLM Benchmark | 52.2% | 72.5% |
| NYT Connections (extended) | 89.6% | 30.1% |
| Chess Puzzles | 33% | 4% |
| LMArena Hard Prompts | 1444 | 1448 |
| Mystery Game Puzzles | 34% | 5% |
| DTBench | 90.9% | 82.1% |
| LMCA | 41.7% | 28.3% |
| Epoch Capabilities Index | 154.49 | 142.38 |
| ARC-AGI-2 | 61.4% | — |
| SimpleBench | 61.1% | — |
| ARC-AGI-1 | 89% | — |
| CritPt | 16.6% | — |
Math DeepSeek V4 Flash leads
DeepSeek V4 Flash: 60.3 (#37), Qwen3 Max: 38.7 (#131)
| Benchmark | DeepSeek V4 Flash | Qwen3 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 57.5% | 18.9% |
| OTIS Mock AIME 2024-2025 | 94.4% | 73.3% |
| LMArena Math | 1427 | 1446 |
| FrontierMath Tier 4 | 24.4% | — |
| MathArena Final-Answer Competitions | 76.5% | — |
| ProofBench | 56% | — |
| MATH Level 5 | — | 97.1% |
Knowledge DeepSeek V4 Flash leads
DeepSeek V4 Flash: 55.4 (#48), Qwen3 Max: 48.1 (#78)
| Benchmark | DeepSeek V4 Flash | Qwen3 Max |
|---|---|---|
| GPQA Diamond | 91% | 72.6% |
| SimpleQA Verified | 33.6% | 48.7% |
| LMArena Expert | 1441 | 1455 |
Multilingual Too close to call
DeepSeek V4 Flash: 53.0 (#72), Qwen3 Max: 53.7 (#62)
| Benchmark | DeepSeek V4 Flash | Qwen3 Max |
|---|---|---|
| LMArena Non-English | 1420 | 1429 |
| LMArena Chinese | 1468 | 1478 |
| LMArena French | 1439 | 1449 |
| LMArena German | 1418 | 1463 |
| LMArena Japanese | 1406 | 1397 |
| LMArena Korean | 1384 | 1399 |
| LMArena Russian | 1428 | 1428 |
| LMArena Spanish | 1436 | 1462 |
Instruction Following Too close to call
DeepSeek V4 Flash: 74.9 (#81), Qwen3 Max: 74.8 (#87)
| Benchmark | DeepSeek V4 Flash | Qwen3 Max |
|---|---|---|
| LMArena Instruction Following | 1421 | 1419 |
Long Context DeepSeek V4 Flash leads
DeepSeek V4 Flash: 43.8 (#85), Qwen3 Max: 41.6 (#134)
| Benchmark | DeepSeek V4 Flash | Qwen3 Max |
|---|---|---|
| LMArena Longer Query | 1434 | 1438 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 14.5% |
Writing & Preference DeepSeek V4 Flash leads
DeepSeek V4 Flash: 63.8 (#61), Qwen3 Max: 62.4 (#76)
| Benchmark | DeepSeek V4 Flash | Qwen3 Max |
|---|---|---|
| LMArena Text | 1432 | 1439 |
| LMArena Creative Writing | 1403 | 1402 |
| LMArena Multi-Turn | 1449 | 1446 |
| EQ-Bench Creative Writing | 1559 | — |
Frequently asked questions
Is DeepSeek V4 Flash better than Qwen3 Max?
DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 43.7 on the Noometry Index.
Which is cheaper, DeepSeek V4 Flash or Qwen3 Max?
DeepSeek V4 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Qwen3 Max lists at $1.20 and $6.
Is DeepSeek V4 Flash or Qwen3 Max better for coding?
DeepSeek V4 Flash scores higher on coding benchmarks: 47.9 versus 43.0 in the Noometry coding category.
Which has the bigger context window?
DeepSeek V4 Flash does, with 1M tokens against 262K.
How many benchmarks do DeepSeek V4 Flash and Qwen3 Max share?
29 benchmarks have published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and Qwen3 Max has 33.