Model comparison
DeepSeek V4.1 Flash vs o1
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 40.9 on the Noometry Index.
Last verified . 24 shared benchmarks.
DeepSeekSummary
- They share 24 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 9 categories and o1 in 1 category; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4.1 Flash leads 66.7 to 36.1.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 67.4% for DeepSeek V4.1 Flash and 14.7% for o1.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $15 / $60 for o1.
- DeepSeek V4.1 Flash accepts more context: 1M tokens versus 200K.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.
Side by side
| DeepSeek V4.1 Flash | o1 | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 52.8 | 40.9 |
| Released | 2026-09-09 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | 1M | 200K |
| Max output | 393K | 100K |
| Input $ / M tokens | $0.15 | $15 |
| Output $ / M tokens | $0.60 | $60 |
| Results tracked | 37 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 52.9 (#32), o1: 46.1 (#70)
| Benchmark | DeepSeek V4.1 Flash | o1 |
|---|---|---|
| LMArena Coding | 1506 | 1367 |
| Aider Polyglot | — | 61.7% |
| LMArena WebDev | 1619 | — |
| SciCode | 51.9% | — |
| WeirdML | — | 47.6% |
| LiveBench Coding | — | 69.7% |
| CadEval | — | 56% |
| ALE-Bench | 1,092 | — |
| HumanEval+ | — | 89% |
| MBPP+ | — | 80.2% |
Agentic & Tool Use DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 31.2 (#69), o1: 24.6 (#117)
| Benchmark | DeepSeek V4.1 Flash | o1 |
|---|---|---|
| APEX-Agents | 39.5% | — |
| Cybench | — | 10% |
| GDP.pdf | 19.8% | — |
| METR Time Horizons | — | 51.1% |
Reasoning DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 50.2 (#36), o1: 27.9 (#111)
| Benchmark | DeepSeek V4.1 Flash | o1 |
|---|---|---|
| LMArena Hard Prompts | 1483 | 1371 |
| DTBench | 89.9% | 74.7% |
| LMCA | 47% | 22.3% |
| Epoch Capabilities Index | 154.9 | 141.91 |
| SimpleBench | — | 41.7% |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | — | 30.7% |
| CritPt | 14.3% | — |
| Chess Puzzles | — | 15% |
| EnigmaEval | — | 5.7% |
| LiveBench Reasoning | — | 91.6% |
| Mystery Game Puzzles | 43% | — |
| LiveBench Data Analysis | — | 65.5% |
| Surface Evolver Bench | 46.3% | — |
| LiveBench | — | 75.7% |
Math DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 66.7 (#25), o1: 36.1 (#175)
| Benchmark | DeepSeek V4.1 Flash | o1 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 67.4% | 14.7% |
| OTIS Mock AIME 2024-2025 | 98.3% | 73.3% |
| LMArena Math | 1477 | 1388 |
| FrontierMath Tier 4 | 26.8% | — |
| ProofBench | 54% | — |
| LiveBench Math | — | 80.3% |
| MATH Level 5 | — | 94.7% |
| FrontierMath (Feb 2025 set) | — | 9.3% |
Knowledge DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 57.9 (#38), o1: 41.5 (#110)
| Benchmark | DeepSeek V4.1 Flash | o1 |
|---|---|---|
| GPQA Diamond | 89.8% | 76.8% |
| LMArena Expert | 1506 | 1361 |
| Humanity's Last Exam | — | 8% |
| SimpleQA Verified | — | 41.1% |
| Confabulations | — | 11.7% |
Multimodal DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 39.1 (#61), o1: 34.2 (#93)
| Benchmark | DeepSeek V4.1 Flash | o1 |
|---|---|---|
| LMArena Vision | 1277 | 1168 |
| GeoBench | — | 80% |
| VPCT | — | 37% |
| Furniture Assembly | 34.2% | — |
| SpatialViz-Bench | — | 41.4% |
Multilingual DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 55.0 (#35), o1: 48.6 (#142)
| Benchmark | DeepSeek V4.1 Flash | o1 |
|---|---|---|
| LMArena Non-English | 1448 | 1358 |
| LMArena Chinese | 1497 | 1394 |
| LMArena French | 1452 | 1344 |
| LMArena German | 1484 | 1337 |
| LMArena Japanese | 1412 | 1346 |
| LMArena Korean | 1452 | 1396 |
| LMArena Russian | 1471 | 1356 |
| LMArena Spanish | 1459 | 1345 |
Instruction Following DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 77.3 (#26), o1: 74.8 (#86)
| Benchmark | DeepSeek V4.1 Flash | o1 |
|---|---|---|
| LMArena Instruction Following | 1474 | 1367 |
| LiveBench Instruction Following | — | 81.5% |
Long Context o1 leads
DeepSeek V4.1 Flash: 45.2 (#47), o1: 50.3 (#9)
| Benchmark | DeepSeek V4.1 Flash | o1 |
|---|---|---|
| LMArena Longer Query | 1475 | 1378 |
| Fiction.LiveBench | — | 83.3% |
Writing & Preference DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 65.4 (#48), o1: 55.6 (#144)
| Benchmark | DeepSeek V4.1 Flash | o1 |
|---|---|---|
| LMArena Text | 1462 | 1366 |
| LMArena Creative Writing | 1435 | 1348 |
| LMArena Multi-Turn | 1457 | 1369 |
| Short-Story Creative Writing | — | 70.2% |
| EQ-Bench Creative Writing | 1540 | — |
| LiveBench Language | — | 65.4% |
Frequently asked questions
Is DeepSeek V4.1 Flash better than o1?
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 40.9 on the Noometry Index.
Which is cheaper, DeepSeek V4.1 Flash or o1?
DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; o1 lists at $15 and $60.
Is DeepSeek V4.1 Flash or o1 better for coding?
DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 46.1 in the Noometry coding category.
Which has the bigger context window?
DeepSeek V4.1 Flash does, with 1M tokens against 200K.
How many benchmarks do DeepSeek V4.1 Flash and o1 share?
24 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and o1 has 52.