Model comparison
DeepSeek-V3.1 vs Qwen3.5 122B-A10B
DeepSeek-V3.1 and Qwen3.5 122B-A10B score almost the same on the Noometry Index (42.8 vs 42.1), so choose on price, context window or the category you care about most.
Last verified . 20 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 5 categories and Qwen3.5 122B-A10B in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in long context, where Qwen3.5 122B-A10B leads 43.0 to 36.3.
- The biggest single-benchmark swing is LMCA: 24.3% for DeepSeek-V3.1 and 32.2% for Qwen3.5 122B-A10B.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $0.40 / $3.20 for Qwen3.5 122B-A10B.
- Qwen3.5 122B-A10B accepts more context: 262K tokens versus 164K.
Side by side
| DeepSeek-V3.1 | Qwen3.5 122B-A10B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 42.8 | 42.1 |
| Released | 2025-08-21 | 2026-02-23 |
| Weights | Open | Open |
| Context window | 164K | 262K |
| Max output | 8K | 66K |
| Input $ / M tokens | $0.25 | $0.40 |
| Output $ / M tokens | $0.95 | $3.20 |
| Results tracked | 27 | 27 |
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), Qwen3.5 122B-A10B: 39.1 (#162)
| Benchmark | DeepSeek-V3.1 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Coding | 1417 | 1436 |
| LMArena WebDev | — | 1360 |
| SciCode | — | 35.6% |
| WeirdML | 38.4% | — |
Reasoning Too close to call
DeepSeek-V3.1: 27.9 (#110), Qwen3.5 122B-A10B: 27.2 (#123)
| Benchmark | DeepSeek-V3.1 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1421 |
| DTBench | 82.7% | 84.3% |
| LMCA | 24.3% | 32.2% |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| NYT Connections (extended) | — | 51.7% |
| CritPt | — | 0.9% |
| Thematic Generalization | — | 51.2% |
| Mystery Game Puzzles | — | 17% |
| Epoch Capabilities Index | 139.92 | — |
| ForecastBench | 58 | — |
Math Too close to call
DeepSeek-V3.1: 38.9 (#122), Qwen3.5 122B-A10B: 39.1 (#112)
| Benchmark | DeepSeek-V3.1 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Math | 1420 | 1432 |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), Qwen3.5 122B-A10B: 38.8 (#142)
| Benchmark | DeepSeek-V3.1 | Qwen3.5 122B-A10B |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | 11.2% |
| LMArena Expert | 1405 | 1432 |
Multimodal Not comparable
DeepSeek-V3.1: —, Qwen3.5 122B-A10B: 39.6 (#57)
| Benchmark | DeepSeek-V3.1 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Vision | — | 1245 |
Multilingual Too close to call
DeepSeek-V3.1: 51.6 (#106), Qwen3.5 122B-A10B: 51.6 (#107)
| Benchmark | DeepSeek-V3.1 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Non-English | 1400 | 1400 |
| LMArena Chinese | 1469 | 1462 |
| LMArena French | 1447 | 1442 |
| LMArena German | 1411 | 1426 |
| LMArena Japanese | 1378 | 1367 |
| LMArena Korean | 1337 | 1352 |
| LMArena Russian | 1405 | 1400 |
| LMArena Spanish | 1431 | 1424 |
Instruction Following Too close to call
DeepSeek-V3.1: 73.9 (#110), Qwen3.5 122B-A10B: 73.8 (#115)
| Benchmark | DeepSeek-V3.1 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Instruction Following | 1400 | 1399 |
Long Context Qwen3.5 122B-A10B leads
DeepSeek-V3.1: 36.3 (#232), Qwen3.5 122B-A10B: 43.0 (#109)
| Benchmark | DeepSeek-V3.1 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Longer Query | 1422 | 1410 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference Too close to call
DeepSeek-V3.1: 60.3 (#98), Qwen3.5 122B-A10B: 60.0 (#105)
| Benchmark | DeepSeek-V3.1 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Text | 1420 | 1417 |
| LMArena Creative Writing | 1401 | 1368 |
| LMArena Multi-Turn | 1408 | 1416 |
| EQ-Bench Creative Writing | 1436 | — |
Frequently asked questions
Is DeepSeek-V3.1 better than Qwen3.5 122B-A10B?
DeepSeek-V3.1 and Qwen3.5 122B-A10B score almost the same on the Noometry Index (42.8 vs 42.1), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-V3.1 or Qwen3.5 122B-A10B?
DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; Qwen3.5 122B-A10B lists at $0.40 and $3.20.
Is DeepSeek-V3.1 or Qwen3.5 122B-A10B better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 39.1 in the Noometry coding category.
Which has the bigger context window?
Qwen3.5 122B-A10B does, with 262K tokens against 164K.
How many benchmarks do DeepSeek-V3.1 and Qwen3.5 122B-A10B share?
20 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Qwen3.5 122B-A10B has 27.