Model comparison
DeepSeek V4 Pro vs Gemma 3 27B
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 30.8 on the Noometry Index. Gemma 3 27B costs 9.9× less per token, which makes it the better buy when DeepSeek V4 Pro's lead doesn't matter for your workload.
Last verified . 28 shared benchmarks.
DeepSeek
GoogleSummary
- They share 28 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 9 categories and Gemma 3 27B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Pro leads 56.5 to 16.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.6% for DeepSeek V4 Pro and 22.5% for Gemma 3 27B.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $0.66 / $1.98 for DeepSeek V4 Pro.
- DeepSeek V4 Pro accepts more context: 1M tokens versus 131K.
Side by side
| DeepSeek V4 Pro | Gemma 3 27B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 54.3 | 30.8 |
| Released | 2026-04-24 | 2025-03-11 |
| Weights | Open | Open |
| Context window | 1M | 131K |
| Max output | 393K | 8K |
| Input $ / M tokens | $0.66 | $0.08 |
| Output $ / M tokens | $1.98 | $0.16 |
| Results tracked | 48 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Pro leads
DeepSeek V4 Pro: 52.4 (#34), Gemma 3 27B: 22.5 (#334)
| Benchmark | DeepSeek V4 Pro | Gemma 3 27B |
|---|---|---|
| SciCode | 51% | 21.2% |
| LMArena Coding | 1470 | 1322 |
| SWE-bench Verified | 77.6% | — |
| FrontierCode | 28.6% | — |
| Aider Polyglot | — | 4.9% |
| LMArena WebDev | 1582 | — |
| WeirdML | 66.2% | — |
| LiveBench Coding | — | 39.9% |
| ALE-Bench | 1,403 | — |
Agentic & Tool Use DeepSeek V4 Pro leads
DeepSeek V4 Pro: 32.8 (#58), Gemma 3 27B: 25.1 (#110)
| Benchmark | DeepSeek V4 Pro | Gemma 3 27B |
|---|---|---|
| APEX-Agents | 47.3% | — |
| Berkeley Function Calling Leaderboard | — | 29.5% |
| Vending-Bench 2 | 3,285 | — |
Reasoning DeepSeek V4 Pro leads
DeepSeek V4 Pro: 56.5 (#24), Gemma 3 27B: 16.7 (#301)
| Benchmark | DeepSeek V4 Pro | Gemma 3 27B |
|---|---|---|
| Kagi LLM Benchmark | 53.5% | 40.4% |
| CritPt | 18% | 0% |
| Chess Puzzles | 47% | 0% |
| LMArena Hard Prompts | 1461 | 1340 |
| DTBench | 93.9% | 52.5% |
| LMCA | 45.5% | 12.3% |
| Epoch Capabilities Index | 155.31 | 130.04 |
| ARC-AGI-2 | 61.3% | — |
| NYT Connections (extended) | 91.3% | — |
| ARC-AGI-1 | 90.5% | — |
| LiveBench Reasoning | — | 43.8% |
| Mystery Game Puzzles | 43% | — |
| LiveBench Data Analysis | — | 51.5% |
| Surface Evolver Bench | 40% | — |
| ForecastBench | 56.1 | — |
| LiveBench | — | 50% |
Math DeepSeek V4 Pro leads
DeepSeek V4 Pro: 64.8 (#30), Gemma 3 27B: 25.9 (#265)
| Benchmark | DeepSeek V4 Pro | Gemma 3 27B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.6% | 22.5% |
| LMArena Math | 1455 | 1312 |
| FrontierMath (Tiers 1-3) | 64.6% | — |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 76.6% | — |
| ProofBench | 50% | — |
| LiveBench Math | — | 55.4% |
| MATH Level 5 | — | 74% |
Knowledge DeepSeek V4 Pro leads
DeepSeek V4 Pro: 59.5 (#31), Gemma 3 27B: 25.5 (#261)
| Benchmark | DeepSeek V4 Pro | Gemma 3 27B |
|---|---|---|
| GPQA Diamond | 91.7% | 47.7% |
| Vectara Hallucination Rate | 8.6% | 7.4% |
| LMArena Expert | 1464 | 1304 |
| SimpleQA Verified | 52.9% | — |
| Confabulations | — | 40.3% |
Multimodal Not comparable
DeepSeek V4 Pro: —, Gemma 3 27B: 32.6 (#100)
| Benchmark | DeepSeek V4 Pro | Gemma 3 27B |
|---|---|---|
| LMArena Vision | — | 1164 |
| GeoBench | — | 52% |
Multilingual DeepSeek V4 Pro leads
DeepSeek V4 Pro: 54.4 (#45), Gemma 3 27B: 46.9 (#155)
| Benchmark | DeepSeek V4 Pro | Gemma 3 27B |
|---|---|---|
| LMArena Non-English | 1439 | 1334 |
| LMArena Chinese | 1486 | 1346 |
| LMArena French | 1472 | 1368 |
| LMArena German | 1458 | 1362 |
| LMArena Japanese | 1445 | 1287 |
| LMArena Korean | 1447 | 1308 |
| LMArena Russian | 1453 | 1349 |
| LMArena Spanish | 1458 | 1349 |
Instruction Following DeepSeek V4 Pro leads
DeepSeek V4 Pro: 76.1 (#47), Gemma 3 27B: 70.6 (#160)
| Benchmark | DeepSeek V4 Pro | Gemma 3 27B |
|---|---|---|
| LMArena Instruction Following | 1448 | 1321 |
| LiveBench Instruction Following | — | 74.9% |
Long Context DeepSeek V4 Pro leads
DeepSeek V4 Pro: 45.0 (#51), Gemma 3 27B: 27.6 (#293)
| Benchmark | DeepSeek V4 Pro | Gemma 3 27B |
|---|---|---|
| LMArena Longer Query | 1458 | 1333 |
| Fiction.LiveBench | — | 33.3% |
| CL-bench Life | 13.5% | — |
Writing & Preference DeepSeek V4 Pro leads
DeepSeek V4 Pro: 65.5 (#46), Gemma 3 27B: 52.5 (#168)
| Benchmark | DeepSeek V4 Pro | Gemma 3 27B |
|---|---|---|
| LMArena Text | 1451 | 1358 |
| LMArena Creative Writing | 1446 | 1346 |
| EQ-Bench Creative Writing | 1553 | 1266 |
| LMArena Multi-Turn | 1467 | 1345 |
| Short-Story Creative Writing | — | 79.9% |
| EQ-Bench 4 | 1166 | — |
| LiveBench Language | — | 34.6% |
Frequently asked questions
Is DeepSeek V4 Pro better than Gemma 3 27B?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 30.8 on the Noometry Index. Gemma 3 27B costs 9.9× less per token, which makes it the better buy when DeepSeek V4 Pro's lead doesn't matter for your workload.
Which is cheaper, DeepSeek V4 Pro 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 V4 Pro lists at $0.66 and $1.98.
Is DeepSeek V4 Pro or Gemma 3 27B better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 22.5 in the Noometry coding category.
Which has the bigger context window?
DeepSeek V4 Pro does, with 1M tokens against 131K.
How many benchmarks do DeepSeek V4 Pro and Gemma 3 27B share?
28 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Gemma 3 27B has 43.