Model comparison
DeepSeek V4.1 Flash vs Gemini 2.5 Flash
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 39.3 on the Noometry Index.
Last verified . 25 shared benchmarks.
DeepSeek
GoogleSummary
- They share 25 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 8 categories and Gemini 2.5 Flash in 2 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4.1 Flash leads 50.2 to 18.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.3% for DeepSeek V4.1 Flash and 73.1% for Gemini 2.5 Flash.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.30 / $2.50 for Gemini 2.5 Flash.
- Gemini 2.5 Flash accepts more context: 1.05M tokens versus 1M.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.
Side by side
| DeepSeek V4.1 Flash | Gemini 2.5 Flash | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 52.8 | 39.3 |
| Released | 2026-09-09 | 2025-04-17 |
| Weights | Open | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 393K | 66K |
| Input $ / M tokens | $0.15 | $0.30 |
| Output $ / M tokens | $0.60 | $2.50 |
| Results tracked | 37 | 54 |
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), Gemini 2.5 Flash: 35.8 (#220)
| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Flash |
|---|---|---|
| LMArena Coding | 1506 | 1424 |
| ALE-Bench | 1,092 | 661.88 |
| SWE-bench Verified (bash only) | — | 28.7% |
| Aider Polyglot | — | 55.1% |
| LMArena WebDev | 1619 | — |
| SciCode | 51.9% | — |
| WeirdML | — | 41.9% |
Agentic & Tool Use Too close to call
DeepSeek V4.1 Flash: 31.2 (#69), Gemini 2.5 Flash: 30.8 (#74)
| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Flash |
|---|---|---|
| Terminal-Bench | — | 17.1% |
| APEX-Agents | 39.5% | — |
| Berkeley Function Calling Leaderboard | — | 56.2% |
| TheAgentCompany | — | 41.1% |
| BALROG | — | 33.5% |
| GDP.pdf | 19.8% | — |
| Vending-Bench 2 | — | 548.84 |
Reasoning DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 50.2 (#36), Gemini 2.5 Flash: 18.1 (#286)
| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Flash |
|---|---|---|
| CritPt | 14.3% | 1.1% |
| LMArena Hard Prompts | 1483 | 1422 |
| DTBench | 89.9% | 76.5% |
| LMCA | 47% | 27.5% |
| Epoch Capabilities Index | 154.9 | 143.03 |
| ARC-AGI-2 | — | 2.5% |
| SimpleBench | — | 41.2% |
| Kagi LLM Benchmark | — | 56.8% |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | — | 33.3% |
| EnigmaEval | — | 2.7% |
| Mystery Game Puzzles | 43% | — |
| Surface Evolver Bench | 46.3% | — |
| ForecastBench | — | 60.6 |
Math DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 66.7 (#25), Gemini 2.5 Flash: 39.9 (#98)
| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.3% | 73.1% |
| LMArena Math | 1477 | 1415 |
| FrontierMath (Tiers 1-3) | 67.4% | — |
| FrontierMath Tier 4 | 26.8% | — |
| ProofBench | 54% | — |
| Omni-MATH | — | 38.5% |
| FrontierMath (Feb 2025 set) | — | 4.8% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 57.9 (#38), Gemini 2.5 Flash: 36.4 (#168)
| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Flash |
|---|---|---|
| LMArena Expert | 1506 | 1426 |
| GPQA Diamond | 89.8% | — |
| Humanity's Last Exam | — | 12.1% |
| MMLU-Pro | — | 63.9% |
| Confabulations | — | 16.8% |
| Vectara Hallucination Rate | — | 7.8% |
| GPQA (HELM) | — | 39% |
Multimodal Gemini 2.5 Flash leads
DeepSeek V4.1 Flash: 39.1 (#61), Gemini 2.5 Flash: 41.8 (#32)
| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Flash |
|---|---|---|
| LMArena Vision | 1277 | 1253 |
| GeoBench | — | 76% |
| VPCT | — | 46.2% |
| Furniture Assembly | 34.2% | — |
| SpatialViz-Bench | — | 36.9% |
Multilingual DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 55.0 (#35), Gemini 2.5 Flash: 52.3 (#88)
| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Flash |
|---|---|---|
| LMArena Non-English | 1448 | 1409 |
| LMArena Chinese | 1497 | 1450 |
| LMArena French | 1452 | 1433 |
| LMArena German | 1484 | 1418 |
| LMArena Japanese | 1412 | 1405 |
| LMArena Korean | 1452 | 1385 |
| LMArena Russian | 1471 | 1415 |
| LMArena Spanish | 1459 | 1421 |
Instruction Following DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 77.3 (#26), Gemini 2.5 Flash: 75.7 (#54)
| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1474 | 1405 |
| IFEval | — | 89.8% |
Long Context Gemini 2.5 Flash leads
DeepSeek V4.1 Flash: 45.2 (#47), Gemini 2.5 Flash: 47.5 (#17)
| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Flash |
|---|---|---|
| LMArena Longer Query | 1475 | 1419 |
| Fiction.LiveBench | — | 77.8% |
Writing & Preference DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 65.4 (#48), Gemini 2.5 Flash: 53.8 (#157)
| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Flash |
|---|---|---|
| LMArena Text | 1462 | 1417 |
| LMArena Creative Writing | 1435 | 1400 |
| EQ-Bench Creative Writing | 1540 | 1137 |
| LMArena Multi-Turn | 1457 | 1408 |
| Short-Story Creative Writing | — | 76.5% |
| WildBench | — | 81.7% |
Frequently asked questions
Is DeepSeek V4.1 Flash better than Gemini 2.5 Flash?
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 39.3 on the Noometry Index.
Which is cheaper, DeepSeek V4.1 Flash or Gemini 2.5 Flash?
DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Gemini 2.5 Flash lists at $0.30 and $2.50.
Is DeepSeek V4.1 Flash or Gemini 2.5 Flash better for coding?
DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 35.8 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Flash does, with 1.05M tokens against 1M.
How many benchmarks do DeepSeek V4.1 Flash and Gemini 2.5 Flash share?
25 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and Gemini 2.5 Flash has 54.