Model comparison
DeepSeek V4 Flash vs Gemma 2 27B
DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 29.4 on the Noometry Index.
Last verified . 22 shared benchmarks.
DeepSeek
GoogleSummary
- They share 22 benchmarks with published results for both. DeepSeek V4 Flash scores higher in 8 categories and Gemma 2 27B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4 Flash leads 60.3 to 10.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 94.4% for DeepSeek V4 Flash and 1.4% for Gemma 2 27B.
- DeepSeek V4 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.65 / $0.65 for Gemma 2 27B.
- DeepSeek V4 Flash accepts more context: 1M tokens versus 8K.
Side by side
| DeepSeek V4 Flash | Gemma 2 27B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 53.6 | 29.4 |
| Released | 2026-04-24 | 2024-06-24 |
| Weights | Open | Open |
| Context window | 1M | 8K |
| Max output | 393K | 2K |
| Input $ / M tokens | $0.15 | $0.65 |
| Output $ / M tokens | $0.60 | $0.65 |
| Results tracked | 41 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Flash leads
DeepSeek V4 Flash: 47.9 (#59), Gemma 2 27B: 34.1 (#246)
| Benchmark | DeepSeek V4 Flash | Gemma 2 27B |
|---|---|---|
| LMArena Coding | 1457 | 1211 |
| FrontierCode | 18.8% | — |
| LMArena WebDev | 1582 | — |
| SciCode | 49.9% | — |
| WeirdML | 63% | — |
| BigCodeBench Instruct | — | 42.8% |
| LiveBench Coding | — | 36% |
| BigCodeBench Complete | — | 52.5% |
| ALE-Bench | 1,306 | — |
Reasoning DeepSeek V4 Flash leads
DeepSeek V4 Flash: 53.7 (#30), Gemma 2 27B: 15.3 (#315)
| Benchmark | DeepSeek V4 Flash | Gemma 2 27B |
|---|---|---|
| LMArena Hard Prompts | 1444 | 1198 |
| DTBench | 90.9% | 48% |
| LMCA | 41.7% | 7.1% |
| Epoch Capabilities Index | 154.49 | 122.08 |
| ARC-AGI-2 | 61.4% | — |
| SimpleBench | 61.1% | — |
| Kagi LLM Benchmark | 52.2% | — |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | 89% | — |
| CritPt | 16.6% | — |
| Chess Puzzles | 33% | — |
| LiveBench Reasoning | — | 28.1% |
| Mystery Game Puzzles | 34% | — |
| LiveBench Data Analysis | — | 47.9% |
| LiveBench | — | 38.2% |
Math DeepSeek V4 Flash leads
DeepSeek V4 Flash: 60.3 (#37), Gemma 2 27B: 10.7 (#311)
| Benchmark | DeepSeek V4 Flash | Gemma 2 27B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 94.4% | 1.4% |
| LMArena Math | 1427 | 1212 |
| FrontierMath (Tiers 1-3) | 57.5% | — |
| FrontierMath Tier 4 | 24.4% | — |
| MathArena Final-Answer Competitions | 76.5% | — |
| ProofBench | 56% | — |
| LiveBench Math | — | 26.5% |
| MATH Level 5 | — | 27.9% |
Knowledge DeepSeek V4 Flash leads
DeepSeek V4 Flash: 55.4 (#48), Gemma 2 27B: 19.0 (#280)
| Benchmark | DeepSeek V4 Flash | Gemma 2 27B |
|---|---|---|
| GPQA Diamond | 91% | 36.5% |
| LMArena Expert | 1441 | 1172 |
| SimpleQA Verified | 33.6% | — |
| Confabulations | — | 27.1% |
| MMLU | — | 75.7% |
Multilingual DeepSeek V4 Flash leads
DeepSeek V4 Flash: 53.0 (#72), Gemma 2 27B: 38.6 (#226)
| Benchmark | DeepSeek V4 Flash | Gemma 2 27B |
|---|---|---|
| LMArena Non-English | 1420 | 1217 |
| LMArena Chinese | 1468 | 1221 |
| LMArena French | 1439 | 1247 |
| LMArena German | 1418 | 1209 |
| LMArena Japanese | 1406 | 1175 |
| LMArena Korean | 1384 | 1174 |
| LMArena Russian | 1428 | 1234 |
| LMArena Spanish | 1436 | 1228 |
Instruction Following DeepSeek V4 Flash leads
DeepSeek V4 Flash: 74.9 (#81), Gemma 2 27B: 60.5 (#249)
| Benchmark | DeepSeek V4 Flash | Gemma 2 27B |
|---|---|---|
| LMArena Instruction Following | 1421 | 1206 |
| LiveBench Instruction Following | — | 58.1% |
Long Context DeepSeek V4 Flash leads
DeepSeek V4 Flash: 43.8 (#85), Gemma 2 27B: 37.3 (#218)
| Benchmark | DeepSeek V4 Flash | Gemma 2 27B |
|---|---|---|
| LMArena Longer Query | 1434 | 1231 |
Writing & Preference DeepSeek V4 Flash leads
DeepSeek V4 Flash: 63.8 (#61), Gemma 2 27B: 44.2 (#225)
| Benchmark | DeepSeek V4 Flash | Gemma 2 27B |
|---|---|---|
| LMArena Text | 1432 | 1231 |
| LMArena Creative Writing | 1403 | 1241 |
| LMArena Multi-Turn | 1449 | 1224 |
| EQ-Bench Creative Writing | 1559 | — |
| LiveBench Language | — | 32.6% |
Frequently asked questions
Is DeepSeek V4 Flash better than Gemma 2 27B?
DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 29.4 on the Noometry Index.
Which is cheaper, DeepSeek V4 Flash or Gemma 2 27B?
DeepSeek V4 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Gemma 2 27B lists at $0.65 and $0.65.
Is DeepSeek V4 Flash or Gemma 2 27B better for coding?
DeepSeek V4 Flash scores higher on coding benchmarks: 47.9 versus 34.1 in the Noometry coding category.
Which has the bigger context window?
DeepSeek V4 Flash does, with 1M tokens against 8K.
How many benchmarks do DeepSeek V4 Flash and Gemma 2 27B share?
22 benchmarks have published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and Gemma 2 27B has 34.