Model comparison
DeepSeek V4.1 Flash vs o3-pro
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 42.9 on the Noometry Index.
Last verified . 3 shared benchmarks.
DeepSeekSummary
- They share 3 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 3 categories and o3-pro in 2 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek V4.1 Flash leads 57.9 to 29.5.
- The biggest single-benchmark swing is LMCA: 47% for DeepSeek V4.1 Flash and 38.5% for o3-pro.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $20 / $80 for o3-pro.
- 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 | o3-pro | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 52.8 | 42.9 |
| Released | 2026-09-09 | 2025-06-10 |
| Weights | Open | Proprietary |
| Context window | 1M | 200K |
| Max output | 393K | 100K |
| Input $ / M tokens | $0.15 | $20 |
| Output $ / M tokens | $0.60 | $80 |
| Results tracked | 37 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-pro leads
DeepSeek V4.1 Flash: 52.9 (#32), o3-pro: 55.5 (#24)
| Benchmark | DeepSeek V4.1 Flash | o3-pro |
|---|---|---|
| Aider Polyglot | — | 84.9% |
| LMArena WebDev | 1619 | — |
| SciCode | 51.9% | — |
| WeirdML | — | 58.2% |
| LMArena Coding | 1506 | — |
| ALE-Bench | 1,092 | — |
Agentic & Tool Use Not comparable
DeepSeek V4.1 Flash: 31.2 (#69), o3-pro: —
| Benchmark | DeepSeek V4.1 Flash | o3-pro |
|---|---|---|
| APEX-Agents | 39.5% | — |
| GDP.pdf | 19.8% | — |
Reasoning DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 50.2 (#36), o3-pro: 23.8 (#171)
| Benchmark | DeepSeek V4.1 Flash | o3-pro |
|---|---|---|
| DTBench | 89.9% | 86.9% |
| LMCA | 47% | 38.5% |
| Epoch Capabilities Index | 154.9 | 147.42 |
| ARC-AGI-2 | — | 4.9% |
| Kagi LLM Benchmark | — | 72.1% |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | — | 59.3% |
| CritPt | 14.3% | — |
| LMArena Hard Prompts | 1483 | — |
| Mystery Game Puzzles | 43% | — |
| Surface Evolver Bench | 46.3% | — |
Math Not comparable
DeepSeek V4.1 Flash: 66.7 (#25), o3-pro: —
| Benchmark | DeepSeek V4.1 Flash | o3-pro |
|---|---|---|
| FrontierMath (Tiers 1-3) | 67.4% | — |
| FrontierMath Tier 4 | 26.8% | — |
| OTIS Mock AIME 2024-2025 | 98.3% | — |
| ProofBench | 54% | — |
| LMArena Math | 1477 | — |
Knowledge DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 57.9 (#38), o3-pro: 29.5 (#238)
| Benchmark | DeepSeek V4.1 Flash | o3-pro |
|---|---|---|
| GPQA Diamond | 89.8% | — |
| Confabulations | — | 14.2% |
| Vectara Hallucination Rate | — | 23.3% |
| LMArena Expert | 1506 | — |
Multimodal Not comparable
DeepSeek V4.1 Flash: 39.1 (#61), o3-pro: —
| Benchmark | DeepSeek V4.1 Flash | o3-pro |
|---|---|---|
| LMArena Vision | 1277 | — |
| Furniture Assembly | 34.2% | — |
Multilingual Not comparable
DeepSeek V4.1 Flash: 55.0 (#35), o3-pro: —
| Benchmark | DeepSeek V4.1 Flash | o3-pro |
|---|---|---|
| LMArena Non-English | 1448 | — |
| LMArena Chinese | 1497 | — |
| LMArena French | 1452 | — |
| LMArena German | 1484 | — |
| LMArena Japanese | 1412 | — |
| LMArena Korean | 1452 | — |
| LMArena Russian | 1471 | — |
| LMArena Spanish | 1459 | — |
Instruction Following Not comparable
DeepSeek V4.1 Flash: 77.3 (#26), o3-pro: —
| Benchmark | DeepSeek V4.1 Flash | o3-pro |
|---|---|---|
| LMArena Instruction Following | 1474 | — |
Long Context o3-pro leads
DeepSeek V4.1 Flash: 45.2 (#47), o3-pro: 72.2 (#1)
| Benchmark | DeepSeek V4.1 Flash | o3-pro |
|---|---|---|
| Fiction.LiveBench | — | 97.2% |
| LMArena Longer Query | 1475 | — |
Writing & Preference DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 65.4 (#48), o3-pro: 57.1 (#133)
| Benchmark | DeepSeek V4.1 Flash | o3-pro |
|---|---|---|
| LMArena Text | 1462 | — |
| LMArena Creative Writing | 1435 | — |
| Short-Story Creative Writing | — | 84.4% |
| EQ-Bench Creative Writing | 1540 | — |
| LMArena Multi-Turn | 1457 | — |
Frequently asked questions
Is DeepSeek V4.1 Flash better than o3-pro?
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 42.9 on the Noometry Index.
Which is cheaper, DeepSeek V4.1 Flash or o3-pro?
DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; o3-pro lists at $20 and $80.
Is DeepSeek V4.1 Flash or o3-pro better for coding?
o3-pro scores higher on coding benchmarks: 55.5 versus 52.9 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 o3-pro share?
3 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and o3-pro has 12.