Model comparison
DeepSeek-V3.1 vs Gemma 3 27B
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 30.8 on the Noometry Index. Gemma 3 27B costs 4.3× less per token, which makes it the better buy when DeepSeek-V3.1's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
DeepSeek
GoogleSummary
- They share 24 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 8 categories and Gemma 3 27B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V3.1 leads 43.7 to 25.5.
- The biggest single-benchmark swing is DTBench: 82.7% for DeepSeek-V3.1 and 52.5% for Gemma 3 27B.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $0.25 / $0.95 for DeepSeek-V3.1.
- DeepSeek-V3.1 accepts more context: 164K tokens versus 131K.
Side by side
| DeepSeek-V3.1 | Gemma 3 27B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 42.8 | 30.8 |
| Released | 2025-08-21 | 2025-03-11 |
| Weights | Open | Open |
| Context window | 164K | 131K |
| Max output | 8K | 8K |
| Input $ / M tokens | $0.25 | $0.08 |
| Output $ / M tokens | $0.95 | $0.16 |
| Results tracked | 27 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1 leads
DeepSeek-V3.1: 40.3 (#144), Gemma 3 27B: 22.5 (#334)
| Benchmark | DeepSeek-V3.1 | Gemma 3 27B |
|---|---|---|
| LMArena Coding | 1417 | 1322 |
| Aider Polyglot | — | 4.9% |
| SciCode | — | 21.2% |
| WeirdML | 38.4% | — |
| LiveBench Coding | — | 39.9% |
Agentic & Tool Use Not comparable
DeepSeek-V3.1: —, Gemma 3 27B: 25.1 (#110)
| Benchmark | DeepSeek-V3.1 | Gemma 3 27B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 29.5% |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.9 (#110), Gemma 3 27B: 16.7 (#301)
| Benchmark | DeepSeek-V3.1 | Gemma 3 27B |
|---|---|---|
| Kagi LLM Benchmark | 53.2% | 40.4% |
| LMArena Hard Prompts | 1417 | 1340 |
| DTBench | 82.7% | 52.5% |
| LMCA | 24.3% | 12.3% |
| Epoch Capabilities Index | 139.92 | 130.04 |
| SimpleBench | 40% | — |
| CritPt | — | 0% |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | — | 43.8% |
| LiveBench Data Analysis | — | 51.5% |
| ForecastBench | 58 | — |
| LiveBench | — | 50% |
Math DeepSeek-V3.1 leads
DeepSeek-V3.1: 38.9 (#122), Gemma 3 27B: 25.9 (#265)
| Benchmark | DeepSeek-V3.1 | Gemma 3 27B |
|---|---|---|
| LMArena Math | 1420 | 1312 |
| OTIS Mock AIME 2024-2025 | — | 22.5% |
| LiveBench Math | — | 55.4% |
| MATH Level 5 | — | 74% |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), Gemma 3 27B: 25.5 (#261)
| Benchmark | DeepSeek-V3.1 | Gemma 3 27B |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | 7.4% |
| LMArena Expert | 1405 | 1304 |
| GPQA Diamond | — | 47.7% |
| Confabulations | — | 40.3% |
Multimodal Not comparable
DeepSeek-V3.1: —, Gemma 3 27B: 32.6 (#100)
| Benchmark | DeepSeek-V3.1 | Gemma 3 27B |
|---|---|---|
| LMArena Vision | — | 1164 |
| GeoBench | — | 52% |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.6 (#106), Gemma 3 27B: 46.9 (#155)
| Benchmark | DeepSeek-V3.1 | Gemma 3 27B |
|---|---|---|
| LMArena Non-English | 1400 | 1334 |
| LMArena Chinese | 1469 | 1346 |
| LMArena French | 1447 | 1368 |
| LMArena German | 1411 | 1362 |
| LMArena Japanese | 1378 | 1287 |
| LMArena Korean | 1337 | 1308 |
| LMArena Russian | 1405 | 1349 |
| LMArena Spanish | 1431 | 1349 |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 73.9 (#110), Gemma 3 27B: 70.6 (#160)
| Benchmark | DeepSeek-V3.1 | Gemma 3 27B |
|---|---|---|
| LMArena Instruction Following | 1400 | 1321 |
| LiveBench Instruction Following | — | 74.9% |
Long Context DeepSeek-V3.1 leads
DeepSeek-V3.1: 36.3 (#232), Gemma 3 27B: 27.6 (#293)
| Benchmark | DeepSeek-V3.1 | Gemma 3 27B |
|---|---|---|
| Fiction.LiveBench | 52.8% | 33.3% |
| LMArena Longer Query | 1422 | 1333 |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.3 (#98), Gemma 3 27B: 52.5 (#168)
| Benchmark | DeepSeek-V3.1 | Gemma 3 27B |
|---|---|---|
| LMArena Text | 1420 | 1358 |
| LMArena Creative Writing | 1401 | 1346 |
| EQ-Bench Creative Writing | 1436 | 1266 |
| LMArena Multi-Turn | 1408 | 1345 |
| Short-Story Creative Writing | — | 79.9% |
| LiveBench Language | — | 34.6% |
Frequently asked questions
Is DeepSeek-V3.1 better than Gemma 3 27B?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 30.8 on the Noometry Index. Gemma 3 27B costs 4.3× less per token, which makes it the better buy when DeepSeek-V3.1's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3.1 or Gemma 3 27B?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; DeepSeek-V3.1 lists at $0.25 and $0.95.
Is DeepSeek-V3.1 or Gemma 3 27B better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 22.5 in the Noometry coding category.
Which has the bigger context window?
DeepSeek-V3.1 does, with 164K tokens against 131K.
How many benchmarks do DeepSeek-V3.1 and Gemma 3 27B share?
24 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Gemma 3 27B has 43.