Model comparison
DeepSeek V4.1 Flash vs Gemini 3.8 Flash
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 52.8 on the Noometry Index. DeepSeek V4.1 Flash costs 5.7× less per token, which makes it the better buy when Gemini 3.8 Flash's lead doesn't matter for your workload.
Last verified . 37 shared benchmarks.
DeepSeek
GoogleSummary
- They share 37 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 1 category and Gemini 3.8 Flash in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.8 Flash leads 76.9 to 50.2.
- The biggest single-benchmark swing is Surface Evolver Bench: 46.3% for DeepSeek V4.1 Flash and 76.9% for Gemini 3.8 Flash.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.75 / $3.75 for Gemini 3.8 Flash.
- Gemini 3.8 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 3.8 Flash | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 52.8 | 61.8 |
| Released | 2026-09-09 | 2026-09-02 |
| Weights | Open | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 393K | 66K |
| Input $ / M tokens | $0.15 | $0.75 |
| Output $ / M tokens | $0.60 | $3.75 |
| Results tracked | 37 | 50 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.8 Flash leads
DeepSeek V4.1 Flash: 52.9 (#32), Gemini 3.8 Flash: 59.2 (#15)
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.8 Flash |
|---|---|---|
| LMArena WebDev | 1619 | 1584 |
| SciCode | 51.9% | 56.6% |
| LMArena Coding | 1506 | 1510 |
| ALE-Bench | 1,092 | 1,270 |
| DeepSWE | — | 73.8% |
| FrontierCode | — | 41.2% |
| CursorBench | — | 39.6% |
| FrontierSWE | — | 19.6% |
| WeirdML | — | 84.8% |
Agentic & Tool Use Gemini 3.8 Flash leads
DeepSeek V4.1 Flash: 31.2 (#69), Gemini 3.8 Flash: 41.8 (#21)
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.8 Flash |
|---|---|---|
| APEX-Agents | 39.5% | 64.3% |
| GDP.pdf | 19.8% | 23.4% |
| Remote Labor Index | — | 5.8% |
| Vending-Bench 2 | — | 5,094 |
Reasoning Gemini 3.8 Flash leads
DeepSeek V4.1 Flash: 50.2 (#36), Gemini 3.8 Flash: 76.9 (#5)
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.8 Flash |
|---|---|---|
| NYT Connections (extended) | 89.6% | 97.4% |
| CritPt | 14.3% | 18.3% |
| LMArena Hard Prompts | 1483 | 1508 |
| Mystery Game Puzzles | 43% | 47% |
| DTBench | 89.9% | 95.7% |
| LMCA | 47% | 52.9% |
| Surface Evolver Bench | 46.3% | 76.9% |
| Epoch Capabilities Index | 154.9 | 156.71 |
| ARC-AGI-2 | — | 89.2% |
| ARC-AGI-1 | — | 98.5% |
| Chess Puzzles | — | 61% |
Math DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 66.7 (#25), Gemini 3.8 Flash: 65.3 (#28)
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.8 Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 67.4% | 68.4% |
| FrontierMath Tier 4 | 26.8% | 22% |
| OTIS Mock AIME 2024-2025 | 98.3% | 98.9% |
| ProofBench | 54% | 48% |
| LMArena Math | 1477 | 1528 |
Knowledge Gemini 3.8 Flash leads
DeepSeek V4.1 Flash: 57.9 (#38), Gemini 3.8 Flash: 74.8 (#2)
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.8 Flash |
|---|---|---|
| GPQA Diamond | 89.8% | 95.4% |
| LMArena Expert | 1506 | 1524 |
| Humanity's Last Exam | — | 44.5% |
| SimpleQA Verified | — | 69.7% |
Multimodal Gemini 3.8 Flash leads
DeepSeek V4.1 Flash: 39.1 (#61), Gemini 3.8 Flash: 40.7 (#45)
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.8 Flash |
|---|---|---|
| LMArena Vision | 1277 | 1314 |
| Furniture Assembly | 34.2% | 31.7% |
| Blueprint-Bench 2 | — | 38.6% |
Multilingual Gemini 3.8 Flash leads
DeepSeek V4.1 Flash: 55.0 (#35), Gemini 3.8 Flash: 58.0 (#5)
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.8 Flash |
|---|---|---|
| LMArena Non-English | 1448 | 1491 |
| LMArena Chinese | 1497 | 1554 |
| LMArena French | 1452 | 1498 |
| LMArena German | 1484 | 1493 |
| LMArena Japanese | 1412 | 1502 |
| LMArena Korean | 1452 | 1459 |
| LMArena Russian | 1471 | 1515 |
| LMArena Spanish | 1459 | 1485 |
Instruction Following Too close to call
DeepSeek V4.1 Flash: 77.3 (#26), Gemini 3.8 Flash: 78.0 (#13)
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.8 Flash |
|---|---|---|
| LMArena Instruction Following | 1474 | 1490 |
Long Context Gemini 3.8 Flash leads
DeepSeek V4.1 Flash: 45.2 (#47), Gemini 3.8 Flash: 46.3 (#24)
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.8 Flash |
|---|---|---|
| LMArena Longer Query | 1475 | 1508 |
Writing & Preference Gemini 3.8 Flash leads
DeepSeek V4.1 Flash: 65.4 (#48), Gemini 3.8 Flash: 72.2 (#15)
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.8 Flash |
|---|---|---|
| LMArena Text | 1462 | 1499 |
| LMArena Creative Writing | 1435 | 1492 |
| EQ-Bench Creative Writing | 1540 | 1748 |
| LMArena Multi-Turn | 1457 | 1501 |
Frequently asked questions
Is DeepSeek V4.1 Flash better than Gemini 3.8 Flash?
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 52.8 on the Noometry Index. DeepSeek V4.1 Flash costs 5.7× less per token, which makes it the better buy when Gemini 3.8 Flash's lead doesn't matter for your workload.
Which is cheaper, DeepSeek V4.1 Flash or Gemini 3.8 Flash?
DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Gemini 3.8 Flash lists at $0.75 and $3.75.
Is DeepSeek V4.1 Flash or Gemini 3.8 Flash better for coding?
Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 52.9 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.8 Flash does, with 1.05M tokens against 1M.
How many benchmarks do DeepSeek V4.1 Flash and Gemini 3.8 Flash share?
37 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and Gemini 3.8 Flash has 50.