Model comparison
DeepSeek-V3.2-Exp vs Qwen3.8 27B
DeepSeek-V3.2-Exp and Qwen3.8 27B score almost the same on the Noometry Index (44.9 vs 45.3), so choose on price, context window or the category you care about most.
Last verified . 26 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 26 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 3 categories and Qwen3.8 27B in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V3.2-Exp leads 51.6 to 40.9.
- The biggest single-benchmark swing is NYT Connections (extended): 36.7% for DeepSeek-V3.2-Exp and 54.5% for Qwen3.8 27B.
- DeepSeek-V3.2-Exp is cheaper at $0.26 / $0.38 per million input/output tokens, against $0.99 / $1.49 for Qwen3.8 27B.
- Qwen3.8 27B accepts more context: 262K tokens versus 164K.
Side by side
| DeepSeek-V3.2-Exp | Qwen3.8 27B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 44.9 | 45.3 |
| Released | 2025-09-29 | 2026-08-14 |
| Weights | Open | Open |
| Context window | 164K | 262K |
| Max output | 66K | 33K |
| Input $ / M tokens | $0.26 | $0.99 |
| Output $ / M tokens | $0.38 | $1.49 |
| Results tracked | 48 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
DeepSeek-V3.2-Exp: 46.4 (#61), Qwen3.8 27B: 50.3 (#42)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 27B |
|---|---|---|
| LMArena WebDev | 1362 | 1592 |
| SciCode | 38.9% | 46.6% |
| LMArena Coding | 1454 | 1487 |
| SWE-bench Verified (bash only) | 70% | — |
| Aider Polyglot | 74.2% | — |
| SWE-bench Multilingual | 59% | — |
| WeirdML | 39.5% | — |
Agentic & Tool Use Not comparable
DeepSeek-V3.2-Exp: 37.9 (#31), Qwen3.8 27B: —
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 27B |
|---|---|---|
| Terminal-Bench | 39.6% | — |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| TheAgentCompany | 42.9% | — |
| Vending-Bench 2 | 1,034 | — |
Reasoning Qwen3.8 27B leads
DeepSeek-V3.2-Exp: 22.0 (#185), Qwen3.8 27B: 31.5 (#71)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 27B |
|---|---|---|
| NYT Connections (extended) | 36.7% | 54.5% |
| CritPt | 2.9% | 5.4% |
| LMArena Hard Prompts | 1433 | 1459 |
| DTBench | 87.7% | 88% |
| LMCA | 29.1% | 41.4% |
| Epoch Capabilities Index | 146.28 | 149.38 |
| ARC-AGI-2 | 4% | — |
| Kagi LLM Benchmark | 52.2% | — |
| ARC-AGI-1 | 57% | — |
| Chess Puzzles | 14% | — |
| Thematic Generalization | 65% | — |
| Surface Evolver Bench | — | 45% |
Math DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 40.8 (#75), Qwen3.8 27B: 35.7 (#139)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 27B |
|---|---|---|
| ProofBench | 8% | 16% |
| LMArena Math | 1435 | 1453 |
| MathArena Final-Answer Competitions | 57.7% | — |
| OTIS Mock AIME 2024-2025 | 87.8% | — |
| FrontierMath (Feb 2025 set) | 22.1% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 51.6 (#61), Qwen3.8 27B: 40.9 (#105)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 27B |
|---|---|---|
| LMArena Expert | 1436 | 1482 |
| GPQA Diamond | 83.4% | — |
| Vectara Hallucination Rate | 5.3% | — |
Multimodal Not comparable
DeepSeek-V3.2-Exp: —, Qwen3.8 27B: 40.5 (#36)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | — | 1270 |
Multilingual Qwen3.8 27B leads
DeepSeek-V3.2-Exp: 52.4 (#85), Qwen3.8 27B: 54.0 (#54)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | 1409 | 1431 |
| LMArena Chinese | 1461 | 1504 |
| LMArena French | 1435 | 1468 |
| LMArena German | 1440 | 1435 |
| LMArena Japanese | 1377 | 1392 |
| LMArena Korean | 1372 | 1389 |
| LMArena Russian | 1423 | 1421 |
| LMArena Spanish | 1441 | 1444 |
Instruction Following Qwen3.8 27B leads
DeepSeek-V3.2-Exp: 74.7 (#89), Qwen3.8 27B: 75.8 (#54)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | 1413 | 1436 |
Long Context DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 47.6 (#15), Qwen3.8 27B: 44.3 (#66)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | 1428 | 1448 |
| Fiction.LiveBench | 83.3% | — |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
Writing & Preference Qwen3.8 27B leads
DeepSeek-V3.2-Exp: 62.5 (#73), Qwen3.8 27B: 65.9 (#41)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 27B |
|---|---|---|
| LMArena Text | 1425 | 1441 |
| LMArena Creative Writing | 1403 | 1382 |
| EQ-Bench Creative Writing | 1515 | 1671 |
| LMArena Multi-Turn | 1427 | 1443 |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than Qwen3.8 27B?
DeepSeek-V3.2-Exp and Qwen3.8 27B score almost the same on the Noometry Index (44.9 vs 45.3), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-V3.2-Exp or Qwen3.8 27B?
DeepSeek-V3.2-Exp is cheaper. It lists at $0.26 per million input tokens and $0.38 per million output tokens; Qwen3.8 27B lists at $0.99 and $1.49.
Is DeepSeek-V3.2-Exp or Qwen3.8 27B better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.3 versus 46.4 in the Noometry coding category.
Which has the bigger context window?
Qwen3.8 27B does, with 262K tokens against 164K.
How many benchmarks do DeepSeek-V3.2-Exp and Qwen3.8 27B share?
26 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 48 scored results on Noometry and Qwen3.8 27B has 28.