Model comparison
DeepSeek-R1 vs Gemma 3 27B
DeepSeek-R1 is the stronger model overall, scoring 42.3 to 30.8 on the Noometry Index. Gemma 3 27B costs 9.1× less per token, which makes it the better buy when DeepSeek-R1'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-R1 scores higher in 9 categories and Gemma 3 27B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-R1 leads 46.3 to 22.5.
- The biggest single-benchmark swing is Aider Polyglot: 71.4% for DeepSeek-R1 and 4.9% for Gemma 3 27B.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $0.50 / $2.15 for DeepSeek-R1.
- DeepSeek-R1 accepts more context: 164K tokens versus 131K.
- Gemma 3 27B has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-R1 | Gemma 3 27B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 42.3 | 30.8 |
| Released | 2025-01-20 | 2025-03-11 |
| Weights | Proprietary | Open |
| Context window | 164K | 131K |
| Max output | 64K | 8K |
| Input $ / M tokens | $0.50 | $0.08 |
| Output $ / M tokens | $2.15 | $0.16 |
| Results tracked | 52 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1 leads
DeepSeek-R1: 46.3 (#68), Gemma 3 27B: 22.5 (#334)
| Benchmark | DeepSeek-R1 | Gemma 3 27B |
|---|---|---|
| Aider Polyglot | 71.4% | 4.9% |
| SciCode | 35.7% | 21.2% |
| LiveBench Coding | 66.7% | 39.9% |
| LMArena Coding | 1427 | 1322 |
| WeirdML | 41.6% | — |
| ALE-Bench | 804.12 | — |
| AlgoTune | 1.7 | — |
Agentic & Tool Use DeepSeek-R1 leads
DeepSeek-R1: 30.7 (#75), Gemma 3 27B: 25.1 (#110)
| Benchmark | DeepSeek-R1 | Gemma 3 27B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 29.5% |
| DeepResearch Bench | 35.1% | — |
| BALROG | 34.9% | — |
| METR Time Horizons | 53.8% | — |
Reasoning DeepSeek-R1 leads
DeepSeek-R1: 18.6 (#278), Gemma 3 27B: 16.7 (#301)
| Benchmark | DeepSeek-R1 | Gemma 3 27B |
|---|---|---|
| Kagi LLM Benchmark | 69.4% | 40.4% |
| CritPt | 1.1% | 0% |
| LiveBench Reasoning | 83.2% | 43.8% |
| LMArena Hard Prompts | 1416 | 1340 |
| LiveBench Data Analysis | 69.8% | 51.5% |
| Epoch Capabilities Index | 141.29 | 130.04 |
| LiveBench | 71.6% | 50% |
| ARC-AGI-2 | 1.3% | — |
| SimpleBench | 40.8% | — |
| ARC-AGI-1 | 21.2% | — |
| Chess Puzzles | — | 0% |
| DTBench | — | 52.5% |
| LMCA | — | 12.3% |
| ForecastBench | 60 | — |
Math DeepSeek-R1 leads
DeepSeek-R1: 43.8 (#79), Gemma 3 27B: 25.9 (#265)
| Benchmark | DeepSeek-R1 | Gemma 3 27B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.4% | 22.5% |
| LiveBench Math | 80.7% | 55.4% |
| LMArena Math | 1400 | 1312 |
| MATH Level 5 | 96.6% | 74% |
| Omni-MATH | 42.4% | — |
Knowledge DeepSeek-R1 leads
DeepSeek-R1: 44.5 (#87), Gemma 3 27B: 25.5 (#261)
| Benchmark | DeepSeek-R1 | Gemma 3 27B |
|---|---|---|
| GPQA Diamond | 76.3% | 47.7% |
| Confabulations | 12.7% | 40.3% |
| Vectara Hallucination Rate | 11.3% | 7.4% |
| LMArena Expert | 1394 | 1304 |
| MMLU-Pro | 79.3% | — |
| GPQA (HELM) | 66.6% | — |
Multimodal Not comparable
DeepSeek-R1: —, Gemma 3 27B: 32.6 (#100)
| Benchmark | DeepSeek-R1 | Gemma 3 27B |
|---|---|---|
| LMArena Vision | — | 1164 |
| GeoBench | — | 52% |
Multilingual DeepSeek-R1 leads
DeepSeek-R1: 52.4 (#85), Gemma 3 27B: 46.9 (#155)
| Benchmark | DeepSeek-R1 | Gemma 3 27B |
|---|---|---|
| LMArena Non-English | 1412 | 1334 |
| LMArena Chinese | 1442 | 1346 |
| LMArena French | 1417 | 1368 |
| LMArena German | 1404 | 1362 |
| LMArena Japanese | 1391 | 1287 |
| LMArena Korean | 1360 | 1308 |
| LMArena Russian | 1423 | 1349 |
| LMArena Spanish | 1411 | 1349 |
Instruction Following DeepSeek-R1 leads
DeepSeek-R1: 72.0 (#143), Gemma 3 27B: 70.6 (#160)
| Benchmark | DeepSeek-R1 | Gemma 3 27B |
|---|---|---|
| LiveBench Instruction Following | 80.5% | 74.9% |
| LMArena Instruction Following | 1382 | 1321 |
| IFEval | 78.4% | — |
Long Context DeepSeek-R1 leads
DeepSeek-R1: 45.4 (#36), Gemma 3 27B: 27.6 (#293)
| Benchmark | DeepSeek-R1 | Gemma 3 27B |
|---|---|---|
| Fiction.LiveBench | 75% | 33.3% |
| LMArena Longer Query | 1391 | 1333 |
Writing & Preference DeepSeek-R1 leads
DeepSeek-R1: 61.4 (#88), Gemma 3 27B: 52.5 (#168)
| Benchmark | DeepSeek-R1 | Gemma 3 27B |
|---|---|---|
| LMArena Text | 1428 | 1358 |
| LMArena Creative Writing | 1405 | 1346 |
| Short-Story Creative Writing | 83% | 79.9% |
| EQ-Bench Creative Writing | 1500 | 1266 |
| LMArena Multi-Turn | 1405 | 1345 |
| LiveBench Language | 48.5% | 34.6% |
| WildBench | 82.8% | — |
Frequently asked questions
Is DeepSeek-R1 better than Gemma 3 27B?
DeepSeek-R1 is the stronger model overall, scoring 42.3 to 30.8 on the Noometry Index. Gemma 3 27B costs 9.1× less per token, which makes it the better buy when DeepSeek-R1's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-R1 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-R1 lists at $0.50 and $2.15.
Is DeepSeek-R1 or Gemma 3 27B better for coding?
DeepSeek-R1 scores higher on coding benchmarks: 46.3 versus 22.5 in the Noometry coding category.
Which has the bigger context window?
DeepSeek-R1 does, with 164K tokens against 131K.
How many benchmarks do DeepSeek-R1 and Gemma 3 27B share?
37 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and Gemma 3 27B has 43.