Model comparison
DeepSeek-V3.1 vs Qwen3.6 35B-A3B
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 37.6 on the Noometry Index.
Last verified . 4 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 4 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 1 category and Qwen3.6 35B-A3B in 3 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.6 35B-A3B leads 51.3 to 43.7.
- The biggest single-benchmark swing is DTBench: 82.7% for DeepSeek-V3.1 and 73.9% for Qwen3.6 35B-A3B.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $0.25 / $1.49 for Qwen3.6 35B-A3B.
- Qwen3.6 35B-A3B accepts more context: 262K tokens versus 164K.
Side by side
| DeepSeek-V3.1 | Qwen3.6 35B-A3B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 42.8 | 37.6 |
| Released | 2025-08-21 | 2026-04-01 |
| Weights | Open | Open |
| Context window | 164K | 262K |
| Max output | 8K | 66K |
| Input $ / M tokens | $0.25 | $0.25 |
| Output $ / M tokens | $0.95 | $1.49 |
| Results tracked | 27 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1 leads
DeepSeek-V3.1: 40.3 (#144), Qwen3.6 35B-A3B: 37.2 (#196)
| Benchmark | DeepSeek-V3.1 | Qwen3.6 35B-A3B |
|---|---|---|
| WeirdML | 38.4% | 34.5% |
| SciCode | — | 35.8% |
| LMArena Coding | 1417 | — |
Agentic & Tool Use Not comparable
DeepSeek-V3.1: —, Qwen3.6 35B-A3B: 22.1 (#134)
| Benchmark | DeepSeek-V3.1 | Qwen3.6 35B-A3B |
|---|---|---|
| Terminal-Bench | — | 23% |
Reasoning Too close to call
DeepSeek-V3.1: 27.9 (#110), Qwen3.6 35B-A3B: 28.0 (#109)
| Benchmark | DeepSeek-V3.1 | Qwen3.6 35B-A3B |
|---|---|---|
| DTBench | 82.7% | 73.9% |
| LMCA | 24.3% | 29.7% |
| Epoch Capabilities Index | 139.92 | 143.93 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| NYT Connections (extended) | — | 41.6% |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 26% |
| LMArena Hard Prompts | 1417 | — |
| Mystery Game Puzzles | — | 22% |
| Surface Evolver Bench | — | 44.4% |
| ForecastBench | 58 | — |
Math Too close to call
DeepSeek-V3.1: 38.9 (#122), Qwen3.6 35B-A3B: 38.9 (#121)
| Benchmark | DeepSeek-V3.1 | Qwen3.6 35B-A3B |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 20.4% |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
| LMArena Math | 1420 | — |
Knowledge Qwen3.6 35B-A3B leads
DeepSeek-V3.1: 43.7 (#90), Qwen3.6 35B-A3B: 51.3 (#68)
| Benchmark | DeepSeek-V3.1 | Qwen3.6 35B-A3B |
|---|---|---|
| GPQA Diamond | — | 84.8% |
| Vectara Hallucination Rate | 5.5% | — |
| LMArena Expert | 1405 | — |
Multilingual Not comparable
DeepSeek-V3.1: 51.6 (#106), Qwen3.6 35B-A3B: —
| Benchmark | DeepSeek-V3.1 | Qwen3.6 35B-A3B |
|---|---|---|
| LMArena Non-English | 1400 | — |
| LMArena Chinese | 1469 | — |
| LMArena French | 1447 | — |
| LMArena German | 1411 | — |
| LMArena Japanese | 1378 | — |
| LMArena Korean | 1337 | — |
| LMArena Russian | 1405 | — |
| LMArena Spanish | 1431 | — |
Instruction Following Not comparable
DeepSeek-V3.1: 73.9 (#110), Qwen3.6 35B-A3B: —
| Benchmark | DeepSeek-V3.1 | Qwen3.6 35B-A3B |
|---|---|---|
| LMArena Instruction Following | 1400 | — |
Long Context Not comparable
DeepSeek-V3.1: 36.3 (#232), Qwen3.6 35B-A3B: —
| Benchmark | DeepSeek-V3.1 | Qwen3.6 35B-A3B |
|---|---|---|
| Fiction.LiveBench | 52.8% | — |
| LMArena Longer Query | 1422 | — |
Writing & Preference Not comparable
DeepSeek-V3.1: 60.3 (#98), Qwen3.6 35B-A3B: —
| Benchmark | DeepSeek-V3.1 | Qwen3.6 35B-A3B |
|---|---|---|
| LMArena Text | 1420 | — |
| LMArena Creative Writing | 1401 | — |
| EQ-Bench Creative Writing | 1436 | — |
| LMArena Multi-Turn | 1408 | — |
Frequently asked questions
Is DeepSeek-V3.1 better than Qwen3.6 35B-A3B?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 37.6 on the Noometry Index.
Which is cheaper, DeepSeek-V3.1 or Qwen3.6 35B-A3B?
DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; Qwen3.6 35B-A3B lists at $0.25 and $1.49.
Is DeepSeek-V3.1 or Qwen3.6 35B-A3B better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 37.2 in the Noometry coding category.
Which has the bigger context window?
Qwen3.6 35B-A3B does, with 262K tokens against 164K.
How many benchmarks do DeepSeek-V3.1 and Qwen3.6 35B-A3B share?
4 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Qwen3.6 35B-A3B has 14.