Model comparison
DeepSeek-V3.2-Exp vs DeepSeek V4.1 Flash
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 44.3 on the Noometry Index.
Last verified . 29 shared benchmarks.
DeepSeekSummary
- They share 29 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 2 categories and DeepSeek V4.1 Flash in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4.1 Flash leads 50.2 to 22.1.
- The biggest single-benchmark swing is NYT Connections (extended): 36.7% for DeepSeek-V3.2-Exp and 89.6% for DeepSeek V4.1 Flash.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.26 / $0.38 for DeepSeek-V3.2-Exp.
- DeepSeek V4.1 Flash accepts more context: 1M tokens versus 164K.
Side by side
| DeepSeek-V3.2-Exp | DeepSeek V4.1 Flash | |
|---|---|---|
| Provider | DeepSeek | DeepSeek |
| Noometry Index | 44.3 | 52.8 |
| Released | 2025-09-29 | 2026-09-09 |
| Weights | Open | Open |
| Context window | 164K | 1M |
| Max output | 66K | 393K |
| Input $ / M tokens | $0.26 | $0.15 |
| Output $ / M tokens | $0.38 | $0.60 |
| Results tracked | 49 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4.1 Flash leads
DeepSeek-V3.2-Exp: 46.5 (#65), DeepSeek V4.1 Flash: 52.9 (#32)
| Benchmark | DeepSeek-V3.2-Exp | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena WebDev | 1362 | 1619 |
| SciCode | 38.9% | 51.9% |
| LMArena Coding | 1454 | 1506 |
| SWE-bench Verified (bash only) | 70% | — |
| Aider Polyglot | 74.2% | — |
| SWE-bench Multilingual | 59% | — |
| WeirdML | 39.5% | — |
| ALE-Bench | — | 1,092 |
Agentic & Tool Use DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 32.7 (#59), DeepSeek V4.1 Flash: 31.2 (#69)
| Benchmark | DeepSeek-V3.2-Exp | DeepSeek V4.1 Flash |
|---|---|---|
| APEX-Agents | 21.3% | 39.5% |
| Terminal-Bench | 39.6% | — |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| TheAgentCompany | 42.9% | — |
| GDP.pdf | — | 19.8% |
| Vending-Bench 2 | 1,034 | — |
Reasoning DeepSeek V4.1 Flash leads
DeepSeek-V3.2-Exp: 22.1 (#208), DeepSeek V4.1 Flash: 50.2 (#36)
| Benchmark | DeepSeek-V3.2-Exp | DeepSeek V4.1 Flash |
|---|---|---|
| NYT Connections (extended) | 36.7% | 89.6% |
| CritPt | 2.9% | 14.3% |
| LMArena Hard Prompts | 1434 | 1483 |
| DTBench | 87.7% | 89.9% |
| LMCA | 29.1% | 47% |
| Epoch Capabilities Index | 146.27 | 154.9 |
| ARC-AGI-2 | 4% | — |
| Kagi LLM Benchmark | 52.2% | — |
| ARC-AGI-1 | 57% | — |
| Chess Puzzles | 14% | — |
| Thematic Generalization | 65% | — |
| Mystery Game Puzzles | — | 43% |
| Surface Evolver Bench | — | 46.3% |
Math DeepSeek V4.1 Flash leads
DeepSeek-V3.2-Exp: 41.7 (#87), DeepSeek V4.1 Flash: 66.7 (#25)
| Benchmark | DeepSeek-V3.2-Exp | DeepSeek V4.1 Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 98.3% |
| ProofBench | 8% | 54% |
| LMArena Math | 1435 | 1477 |
| FrontierMath (Tiers 1-3) | — | 67.4% |
| FrontierMath Tier 4 | — | 26.8% |
| MathArena Final-Answer Competitions | 57.7% | — |
| FrontierMath (Feb 2025 set) | 22.1% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge DeepSeek V4.1 Flash leads
DeepSeek-V3.2-Exp: 51.7 (#66), DeepSeek V4.1 Flash: 57.9 (#38)
| Benchmark | DeepSeek-V3.2-Exp | DeepSeek V4.1 Flash |
|---|---|---|
| GPQA Diamond | 83.4% | 89.8% |
| LMArena Expert | 1436 | 1506 |
| Vectara Hallucination Rate | 5.3% | — |
Multimodal Not comparable
DeepSeek-V3.2-Exp: —, DeepSeek V4.1 Flash: 39.1 (#61)
| Benchmark | DeepSeek-V3.2-Exp | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Vision | — | 1277 |
| Furniture Assembly | — | 34.2% |
Multilingual DeepSeek V4.1 Flash leads
DeepSeek-V3.2-Exp: 52.2 (#90), DeepSeek V4.1 Flash: 55.0 (#35)
| Benchmark | DeepSeek-V3.2-Exp | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Non-English | 1409 | 1448 |
| LMArena Chinese | 1461 | 1497 |
| LMArena French | 1433 | 1452 |
| LMArena German | 1440 | 1484 |
| LMArena Japanese | 1374 | 1412 |
| LMArena Korean | 1371 | 1452 |
| LMArena Russian | 1424 | 1471 |
| LMArena Spanish | 1440 | 1459 |
Instruction Following DeepSeek V4.1 Flash leads
DeepSeek-V3.2-Exp: 74.5 (#93), DeepSeek V4.1 Flash: 77.3 (#26)
| Benchmark | DeepSeek-V3.2-Exp | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Instruction Following | 1413 | 1474 |
Long Context DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 47.6 (#16), DeepSeek V4.1 Flash: 45.2 (#47)
| Benchmark | DeepSeek-V3.2-Exp | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Longer Query | 1428 | 1475 |
| Fiction.LiveBench | 83.3% | — |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
Writing & Preference DeepSeek V4.1 Flash leads
DeepSeek-V3.2-Exp: 62.4 (#77), DeepSeek V4.1 Flash: 65.4 (#48)
| Benchmark | DeepSeek-V3.2-Exp | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Text | 1425 | 1462 |
| LMArena Creative Writing | 1403 | 1435 |
| EQ-Bench Creative Writing | 1515 | 1540 |
| LMArena Multi-Turn | 1427 | 1457 |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than DeepSeek V4.1 Flash?
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 44.3 on the Noometry Index.
Which is cheaper, DeepSeek-V3.2-Exp or DeepSeek V4.1 Flash?
DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; DeepSeek-V3.2-Exp lists at $0.26 and $0.38.
Is DeepSeek-V3.2-Exp or DeepSeek V4.1 Flash better for coding?
DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 46.5 in the Noometry coding category.
Which has the bigger context window?
DeepSeek V4.1 Flash does, with 1M tokens against 164K.
How many benchmarks do DeepSeek-V3.2-Exp and DeepSeek V4.1 Flash share?
29 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 49 scored results on Noometry and DeepSeek V4.1 Flash has 37.