Model comparison
DeepSeek V4 Pro vs Gemma 7B
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 30.0 on the Noometry Index.
Last verified . 14 shared benchmarks.
DeepSeek
GoogleSummary
- They share 14 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 8 categories and Gemma 7B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek V4 Pro leads 65.5 to 27.1.
Side by side
| DeepSeek V4 Pro | Gemma 7B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 54.3 | 30.0 |
| Released | 2026-04-24 | 2024-02-21 |
| Weights | Open | Open |
| Context window | 1M | — |
| Max output | 393K | — |
| Input $ / M tokens | $0.66 | — |
| Output $ / M tokens | $1.98 | — |
| Results tracked | 48 | 27 |
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 7B: 30.5 (#294)
| Benchmark | DeepSeek V4 Pro | Gemma 7B |
|---|---|---|
| LMArena Coding | 1470 | 1048 |
| SWE-bench Verified | 77.6% | — |
| FrontierCode | 28.6% | — |
| LMArena WebDev | 1582 | — |
| SciCode | 51% | — |
| WeirdML | 66.2% | — |
| ALE-Bench | 1,403 | — |
| HumanEval+ | — | 28.7% |
| MBPP+ | — | 43.4% |
Agentic & Tool Use Not comparable
DeepSeek V4 Pro: 32.8 (#58), Gemma 7B: —
| Benchmark | DeepSeek V4 Pro | Gemma 7B |
|---|---|---|
| APEX-Agents | 47.3% | — |
| Vending-Bench 2 | 3,285 | — |
Reasoning DeepSeek V4 Pro leads
DeepSeek V4 Pro: 56.5 (#24), Gemma 7B: 19.9 (#249)
| Benchmark | DeepSeek V4 Pro | Gemma 7B |
|---|---|---|
| LMArena Hard Prompts | 1461 | 1042 |
| Epoch Capabilities Index | 155.31 | 111.99 |
| ARC-AGI-2 | 61.3% | — |
| Kagi LLM Benchmark | 53.5% | — |
| 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% | — |
| Adversarial NLI | — | 48.7% |
| BIG-Bench Hard | — | 55.1% |
| ForecastBench | 56.1 | — |
| HellaSwag | — | 82.2% |
| PIQA | — | 81.2% |
| WinoGrande | — | 79% |
Math DeepSeek V4 Pro leads
DeepSeek V4 Pro: 64.8 (#30), Gemma 7B: 31.2 (#228)
| Benchmark | DeepSeek V4 Pro | Gemma 7B |
|---|---|---|
| LMArena Math | 1455 | 1066 |
| 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% | — |
| GSM8K | — | 46.4% |
Knowledge DeepSeek V4 Pro leads
DeepSeek V4 Pro: 59.5 (#31), Gemma 7B: 27.3 (#252)
| Benchmark | DeepSeek V4 Pro | Gemma 7B |
|---|---|---|
| LMArena Expert | 1464 | 1001 |
| GPQA Diamond | 91.7% | — |
| SimpleQA Verified | 52.9% | — |
| Vectara Hallucination Rate | 8.6% | — |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 83.2% |
| MMLU | — | 66.1% |
| OpenBookQA | — | 78.6% |
| TriviaQA | — | 72.3% |
Multilingual DeepSeek V4 Pro leads
DeepSeek V4 Pro: 54.4 (#45), Gemma 7B: 25.1 (#287)
| Benchmark | DeepSeek V4 Pro | Gemma 7B |
|---|---|---|
| LMArena Non-English | 1439 | 999 |
| LMArena Chinese | 1486 | 1035 |
| LMArena French | 1472 | 1025 |
| LMArena Russian | 1453 | 993 |
| LMArena German | 1458 | — |
| LMArena Japanese | 1445 | — |
| LMArena Korean | 1447 | — |
| LMArena Spanish | 1458 | — |
Instruction Following DeepSeek V4 Pro leads
DeepSeek V4 Pro: 76.1 (#47), Gemma 7B: 51.5 (#295)
| Benchmark | DeepSeek V4 Pro | Gemma 7B |
|---|---|---|
| LMArena Instruction Following | 1448 | 1017 |
Long Context DeepSeek V4 Pro leads
DeepSeek V4 Pro: 45.0 (#51), Gemma 7B: 31.1 (#282)
| Benchmark | DeepSeek V4 Pro | Gemma 7B |
|---|---|---|
| LMArena Longer Query | 1458 | 1022 |
| CL-bench Life | 13.5% | — |
Writing & Preference DeepSeek V4 Pro leads
DeepSeek V4 Pro: 65.5 (#46), Gemma 7B: 27.1 (#302)
| Benchmark | DeepSeek V4 Pro | Gemma 7B |
|---|---|---|
| LMArena Text | 1451 | 1056 |
| LMArena Creative Writing | 1446 | 1024 |
| LMArena Multi-Turn | 1467 | 963 |
| EQ-Bench Creative Writing | 1553 | — |
| EQ-Bench 4 | 1166 | — |
Frequently asked questions
Is DeepSeek V4 Pro better than Gemma 7B?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 30.0 on the Noometry Index.
Is DeepSeek V4 Pro or Gemma 7B better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 30.5 in the Noometry coding category.
How many benchmarks do DeepSeek V4 Pro and Gemma 7B share?
14 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Gemma 7B has 27.