Model comparison
DeepSeek V4 Pro vs Gemma 3n E4b IT
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 37.3 on the Noometry Index.
Last verified . 18 shared benchmarks.
DeepSeek
GoogleSummary
- They share 18 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 8 categories and Gemma 3n E4b IT in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Pro leads 56.5 to 19.9.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 53.5% for DeepSeek V4 Pro and 31.5% for Gemma 3n E4b IT.
Side by side
| DeepSeek V4 Pro | Gemma 3n E4b IT | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 54.3 | 37.3 |
| Released | 2026-04-24 | — |
| Weights | Open | Open |
| Context window | 1M | — |
| Max output | 393K | — |
| Input $ / M tokens | $0.66 | — |
| Output $ / M tokens | $1.98 | — |
| Results tracked | 48 | 18 |
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 3n E4b IT: 37.0 (#198)
| Benchmark | DeepSeek V4 Pro | Gemma 3n E4b IT |
|---|---|---|
| LMArena Coding | 1470 | 1268 |
| SWE-bench Verified | 77.6% | — |
| FrontierCode | 28.6% | — |
| LMArena WebDev | 1582 | — |
| SciCode | 51% | — |
| WeirdML | 66.2% | — |
| ALE-Bench | 1,403 | — |
Agentic & Tool Use Not comparable
DeepSeek V4 Pro: 32.8 (#58), Gemma 3n E4b IT: —
| Benchmark | DeepSeek V4 Pro | Gemma 3n E4b IT |
|---|---|---|
| APEX-Agents | 47.3% | — |
| Vending-Bench 2 | 3,285 | — |
Reasoning DeepSeek V4 Pro leads
DeepSeek V4 Pro: 56.5 (#24), Gemma 3n E4b IT: 19.9 (#247)
| Benchmark | DeepSeek V4 Pro | Gemma 3n E4b IT |
|---|---|---|
| Kagi LLM Benchmark | 53.5% | 31.5% |
| LMArena Hard Prompts | 1461 | 1284 |
| ARC-AGI-2 | 61.3% | — |
| NYT Connections (extended) | 91.3% | — |
| ARC-AGI-1 | 90.5% | — |
| CritPt | 18% | — |
| Chess Puzzles | 47% | — |
| Mystery Game Puzzles | 43% | — |
| DTBench | 93.9% | — |
| LMCA | 45.5% | — |
| Surface Evolver Bench | 40% | — |
| Epoch Capabilities Index | 155.31 | — |
| ForecastBench | 56.1 | — |
Math DeepSeek V4 Pro leads
DeepSeek V4 Pro: 64.8 (#30), Gemma 3n E4b IT: 35.1 (#188)
| Benchmark | DeepSeek V4 Pro | Gemma 3n E4b IT |
|---|---|---|
| LMArena Math | 1455 | 1251 |
| FrontierMath (Tiers 1-3) | 64.6% | — |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 76.6% | — |
| OTIS Mock AIME 2024-2025 | 98.6% | — |
| ProofBench | 50% | — |
Knowledge DeepSeek V4 Pro leads
DeepSeek V4 Pro: 59.5 (#31), Gemma 3n E4b IT: 34.2 (#198)
| Benchmark | DeepSeek V4 Pro | Gemma 3n E4b IT |
|---|---|---|
| LMArena Expert | 1464 | 1246 |
| GPQA Diamond | 91.7% | — |
| SimpleQA Verified | 52.9% | — |
| Vectara Hallucination Rate | 8.6% | — |
Multilingual DeepSeek V4 Pro leads
DeepSeek V4 Pro: 54.4 (#45), Gemma 3n E4b IT: 43.4 (#183)
| Benchmark | DeepSeek V4 Pro | Gemma 3n E4b IT |
|---|---|---|
| LMArena Non-English | 1439 | 1285 |
| LMArena Chinese | 1486 | 1309 |
| LMArena French | 1472 | 1330 |
| LMArena German | 1458 | 1311 |
| LMArena Japanese | 1445 | 1272 |
| LMArena Korean | 1447 | 1259 |
| LMArena Russian | 1453 | 1288 |
| LMArena Spanish | 1458 | 1305 |
Instruction Following DeepSeek V4 Pro leads
DeepSeek V4 Pro: 76.1 (#47), Gemma 3n E4b IT: 66.1 (#210)
| Benchmark | DeepSeek V4 Pro | Gemma 3n E4b IT |
|---|---|---|
| LMArena Instruction Following | 1448 | 1255 |
Long Context DeepSeek V4 Pro leads
DeepSeek V4 Pro: 45.0 (#51), Gemma 3n E4b IT: 38.7 (#191)
| Benchmark | DeepSeek V4 Pro | Gemma 3n E4b IT |
|---|---|---|
| LMArena Longer Query | 1458 | 1276 |
| CL-bench Life | 13.5% | — |
Writing & Preference DeepSeek V4 Pro leads
DeepSeek V4 Pro: 65.5 (#46), Gemma 3n E4b IT: 50.1 (#186)
| Benchmark | DeepSeek V4 Pro | Gemma 3n E4b IT |
|---|---|---|
| LMArena Text | 1451 | 1306 |
| LMArena Creative Writing | 1446 | 1287 |
| LMArena Multi-Turn | 1467 | 1276 |
| EQ-Bench Creative Writing | 1553 | — |
| EQ-Bench 4 | 1166 | — |
Frequently asked questions
Is DeepSeek V4 Pro better than Gemma 3n E4b IT?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 37.3 on the Noometry Index.
Is DeepSeek V4 Pro or Gemma 3n E4b IT better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 37.0 in the Noometry coding category.
How many benchmarks do DeepSeek V4 Pro and Gemma 3n E4b IT share?
18 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Gemma 3n E4b IT has 18.