Model comparison
DeepSeek-V3 vs Qwen3-Coder 480B-A35B Instruct
DeepSeek-V3 is the stronger model overall, scoring 39.5 to 38.1 on the Noometry Index.
Last verified . 19 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 19 benchmarks with published results for both. DeepSeek-V3 scores higher in 5 categories and Qwen3-Coder 480B-A35B Instruct in 3 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in long context, where Qwen3-Coder 480B-A35B Instruct leads 42.0 to 34.0.
- The biggest single-benchmark swing is WeirdML: 36.1% for DeepSeek-V3 and 41.2% for Qwen3-Coder 480B-A35B Instruct.
- DeepSeek-V3 is cheaper at $0.24 / $0.90 per million input/output tokens, against $1.50 / $7.50 for Qwen3-Coder 480B-A35B Instruct.
- Qwen3-Coder 480B-A35B Instruct accepts more context: 262K tokens versus 164K.
Side by side
| DeepSeek-V3 | Qwen3-Coder 480B-A35B Instruct | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 39.5 | 38.1 |
| Released | 2024-12-26 | 2025-04 |
| Weights | Open | Open |
| Context window | 164K | 262K |
| Max output | 164K | 66K |
| Input $ / M tokens | $0.24 | $1.50 |
| Output $ / M tokens | $0.90 | $7.50 |
| Results tracked | 60 | 25 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3 leads
DeepSeek-V3: 42.3 (#106), Qwen3-Coder 480B-A35B Instruct: 35.5 (#223)
| Benchmark | DeepSeek-V3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| WeirdML | 36.1% | 41.2% |
| LMArena Coding | 1368 | 1412 |
| SWE-bench Verified (bash only) | — | 55.4% |
| Aider Polyglot | 55.1% | — |
| LMArena WebDev | — | 1275 |
| SciCode | 35.8% | — |
| GSO | — | 4.9% |
| BigCodeBench Instruct | 50% | — |
| LiveBench Coding | 70.9% | — |
| BigCodeBench Complete | 62.2% | — |
| ALE-Bench | — | 461.45 |
| AlgoTune | — | 1.44 |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73% | — |
Agentic & Tool Use Not comparable
DeepSeek-V3: —, Qwen3-Coder 480B-A35B Instruct: 23.9 (#123)
| Benchmark | DeepSeek-V3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Terminal-Bench | — | 27.2% |
| METR Time Horizons | 49.6% | — |
Reasoning Qwen3-Coder 480B-A35B Instruct leads
DeepSeek-V3: 20.5 (#236), Qwen3-Coder 480B-A35B Instruct: 25.5 (#149)
| Benchmark | DeepSeek-V3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Kagi LLM Benchmark | 52.3% | 49.5% |
| LMArena Hard Prompts | 1365 | 1372 |
| SimpleBench | 27.2% | — |
| 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-Coder 480B-A35B Instruct leads
DeepSeek-V3: 32.1 (#219), Qwen3-Coder 480B-A35B Instruct: 37.6 (#150)
| Benchmark | DeepSeek-V3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Math | 1373 | 1365 |
| 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 Too close to call
DeepSeek-V3: 37.5 (#155), Qwen3-Coder 480B-A35B Instruct: 37.0 (#162)
| Benchmark | DeepSeek-V3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Expert | 1351 | 1338 |
| 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% | — |
Multilingual Too close to call
DeepSeek-V3: 48.5 (#143), Qwen3-Coder 480B-A35B Instruct: 47.7 (#148)
| Benchmark | DeepSeek-V3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Non-English | 1358 | 1346 |
| LMArena Chinese | 1391 | 1357 |
| LMArena French | 1385 | 1398 |
| LMArena German | 1374 | 1325 |
| LMArena Japanese | 1333 | 1310 |
| LMArena Korean | 1319 | 1305 |
| LMArena Russian | 1373 | 1366 |
| LMArena Spanish | 1358 | 1360 |
Instruction Following DeepSeek-V3 leads
DeepSeek-V3: 72.8 (#130), Qwen3-Coder 480B-A35B Instruct: 71.6 (#147)
| Benchmark | DeepSeek-V3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Instruction Following | 1345 | 1355 |
| LiveBench Instruction Following | 81.5% | — |
| IFEval | 83.2% | — |
Long Context Qwen3-Coder 480B-A35B Instruct leads
DeepSeek-V3: 34.0 (#253), Qwen3-Coder 480B-A35B Instruct: 42.0 (#131)
| Benchmark | DeepSeek-V3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Longer Query | 1352 | 1378 |
| Fiction.LiveBench | 50% | — |
Writing & Preference DeepSeek-V3 leads
DeepSeek-V3: 57.4 (#130), Qwen3-Coder 480B-A35B Instruct: 55.3 (#147)
| Benchmark | DeepSeek-V3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Text | 1375 | 1357 |
| LMArena Creative Writing | 1364 | 1333 |
| LMArena Multi-Turn | 1389 | 1365 |
| 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-Coder 480B-A35B Instruct?
DeepSeek-V3 is the stronger model overall, scoring 39.5 to 38.1 on the Noometry Index.
Which is cheaper, DeepSeek-V3 or Qwen3-Coder 480B-A35B Instruct?
DeepSeek-V3 is cheaper. It lists at $0.24 per million input tokens and $0.90 per million output tokens; Qwen3-Coder 480B-A35B Instruct lists at $1.50 and $7.50.
Is DeepSeek-V3 or Qwen3-Coder 480B-A35B Instruct better for coding?
DeepSeek-V3 scores higher on coding benchmarks: 42.3 versus 35.5 in the Noometry coding category.
Which has the bigger context window?
Qwen3-Coder 480B-A35B Instruct does, with 262K tokens against 164K.
How many benchmarks do DeepSeek-V3 and Qwen3-Coder 480B-A35B Instruct share?
19 benchmarks have published results for both models. DeepSeek-V3 has 60 scored results on Noometry and Qwen3-Coder 480B-A35B Instruct has 25.