Model comparison
DeepSeek-V3 vs Qwen3-VL 235B-A22B
Qwen3-VL 235B-A22B is the stronger model overall, scoring 43.2 to 39.5 on the Noometry Index. DeepSeek-V3 costs 3.0× less per token, which makes it the better buy when Qwen3-VL 235B-A22B's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 17 benchmarks with published results for both. DeepSeek-V3 scores higher in 0 categories and Qwen3-VL 235B-A22B in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in long context, where Qwen3-VL 235B-A22B leads 43.4 to 34.0.
- DeepSeek-V3 is cheaper at $0.24 / $0.90 per million input/output tokens, against $0.70 / $2.80 for Qwen3-VL 235B-A22B.
- DeepSeek-V3 accepts more context: 164K tokens versus 131K.
Side by side
| DeepSeek-V3 | Qwen3-VL 235B-A22B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 39.5 | 43.2 |
| Released | 2024-12-26 | 2025-04 |
| Weights | Open | Open |
| Context window | 164K | 131K |
| Max output | 164K | 33K |
| Input $ / M tokens | $0.24 | $0.70 |
| Output $ / M tokens | $0.90 | $2.80 |
| Results tracked | 60 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek-V3: 42.3 (#106), Qwen3-VL 235B-A22B: 42.4 (#100)
| Benchmark | DeepSeek-V3 | Qwen3-VL 235B-A22B |
|---|---|---|
| LMArena Coding | 1368 | 1439 |
| Aider Polyglot | 55.1% | — |
| SciCode | 35.8% | — |
| WeirdML | 36.1% | — |
| BigCodeBench Instruct | 50% | — |
| LiveBench Coding | 70.9% | — |
| BigCodeBench Complete | 62.2% | — |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73% | — |
Agentic & Tool Use Not comparable
DeepSeek-V3: —, Qwen3-VL 235B-A22B: —
| Benchmark | DeepSeek-V3 | Qwen3-VL 235B-A22B |
|---|---|---|
| METR Time Horizons | 49.6% | — |
Reasoning Qwen3-VL 235B-A22B leads
DeepSeek-V3: 20.5 (#236), Qwen3-VL 235B-A22B: 29.3 (#92)
| Benchmark | DeepSeek-V3 | Qwen3-VL 235B-A22B |
|---|---|---|
| LMArena Hard Prompts | 1365 | 1428 |
| SimpleBench | 27.2% | — |
| Kagi LLM Benchmark | 52.3% | — |
| CritPt | 0% | — |
| LiveBench Reasoning | 65.8% | — |
| DTBench | 64.8% | — |
| LiveBench Data Analysis | 60.9% | — |
| LMCA | 15.5% | — |
| BIG-Bench Hard | 87.5% | — |
| Epoch Capabilities Index | 135.94 | — |
| ForecastBench | 59.1 | — |
| HellaSwag | 88.9% | — |
| LiveBench | 66.9% | — |
| PIQA | 84.7% | — |
| WinoGrande | 85.2% | — |
Math Qwen3-VL 235B-A22B leads
DeepSeek-V3: 32.1 (#219), Qwen3-VL 235B-A22B: 39.0 (#118)
| Benchmark | DeepSeek-V3 | Qwen3-VL 235B-A22B |
|---|---|---|
| LMArena Math | 1373 | 1426 |
| OTIS Mock AIME 2024-2025 | 37.8% | — |
| Omni-MATH | 40.3% | — |
| LiveBench Math | 73.5% | — |
| MATH Level 5 | 75.5% | — |
| FrontierMath (Feb 2025 set) | 1.7% | — |
Knowledge Qwen3-VL 235B-A22B leads
DeepSeek-V3: 37.5 (#155), Qwen3-VL 235B-A22B: 40.3 (#121)
| Benchmark | DeepSeek-V3 | Qwen3-VL 235B-A22B |
|---|---|---|
| LMArena Expert | 1351 | 1442 |
| GPQA Diamond | 67.6% | — |
| 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% | — |
Multimodal Not comparable
DeepSeek-V3: —, Qwen3-VL 235B-A22B: 39.8 (#55)
| Benchmark | DeepSeek-V3 | Qwen3-VL 235B-A22B |
|---|---|---|
| LMArena Vision | — | 1247 |
Multilingual Qwen3-VL 235B-A22B leads
DeepSeek-V3: 48.5 (#143), Qwen3-VL 235B-A22B: 51.9 (#97)
| Benchmark | DeepSeek-V3 | Qwen3-VL 235B-A22B |
|---|---|---|
| LMArena Non-English | 1358 | 1405 |
| LMArena Chinese | 1391 | 1463 |
| LMArena French | 1385 | 1452 |
| LMArena German | 1374 | 1424 |
| LMArena Japanese | 1333 | 1385 |
| LMArena Korean | 1319 | 1394 |
| LMArena Russian | 1373 | 1408 |
| LMArena Spanish | 1358 | 1428 |
Instruction Following Qwen3-VL 235B-A22B leads
DeepSeek-V3: 72.8 (#130), Qwen3-VL 235B-A22B: 74.2 (#101)
| Benchmark | DeepSeek-V3 | Qwen3-VL 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1345 | 1406 |
| LiveBench Instruction Following | 81.5% | — |
| IFEval | 83.2% | — |
Long Context Qwen3-VL 235B-A22B leads
DeepSeek-V3: 34.0 (#253), Qwen3-VL 235B-A22B: 43.4 (#98)
| Benchmark | DeepSeek-V3 | Qwen3-VL 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1352 | 1420 |
| Fiction.LiveBench | 50% | — |
Writing & Preference Qwen3-VL 235B-A22B leads
DeepSeek-V3: 57.4 (#130), Qwen3-VL 235B-A22B: 60.2 (#99)
| Benchmark | DeepSeek-V3 | Qwen3-VL 235B-A22B |
|---|---|---|
| LMArena Text | 1375 | 1420 |
| LMArena Creative Writing | 1364 | 1366 |
| LMArena Multi-Turn | 1389 | 1428 |
| 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-VL 235B-A22B?
Qwen3-VL 235B-A22B is the stronger model overall, scoring 43.2 to 39.5 on the Noometry Index. DeepSeek-V3 costs 3.0× less per token, which makes it the better buy when Qwen3-VL 235B-A22B's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3 or Qwen3-VL 235B-A22B?
DeepSeek-V3 is cheaper. It lists at $0.24 per million input tokens and $0.90 per million output tokens; Qwen3-VL 235B-A22B lists at $0.70 and $2.80.
Is DeepSeek-V3 or Qwen3-VL 235B-A22B better for coding?
They score almost the same on coding (42.3 vs 42.4); test both on your own repository before choosing.
Which has the bigger context window?
DeepSeek-V3 does, with 164K tokens against 131K.
How many benchmarks do DeepSeek-V3 and Qwen3-VL 235B-A22B share?
17 benchmarks have published results for both models. DeepSeek-V3 has 60 scored results on Noometry and Qwen3-VL 235B-A22B has 18.