Model comparison
DeepSeek LLM 67B vs Gemma 3 4B
Gemma 3 4B is the stronger model overall, scoring 28.1 to 24.9 on the Noometry Index.
Last verified . 13 shared benchmarks.
DeepSeek
GoogleSummary
- They share 13 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 1 category and Gemma 3 4B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Gemma 3 4B leads 42.5 to 29.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.8% for DeepSeek LLM 67B and 7.5% for Gemma 3 4B.
Side by side
| DeepSeek LLM 67B | Gemma 3 4B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 24.9 | 28.1 |
| Released | 2023-11-29 | 2025-03-12 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.04 |
| Output $ / M tokens | — | $0.08 |
| Results tracked | 15 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 3 4B leads
DeepSeek LLM 67B: 31.9 (#278), Gemma 3 4B: 35.9 (#215)
| Benchmark | DeepSeek LLM 67B | Gemma 3 4B |
|---|---|---|
| LMArena Coding | 1096 | 1230 |
Agentic & Tool Use Not comparable
DeepSeek LLM 67B: —, Gemma 3 4B: 20.9 (#142)
| Benchmark | DeepSeek LLM 67B | Gemma 3 4B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 19.6% |
Reasoning DeepSeek LLM 67B leads
DeepSeek LLM 67B: 16.5 (#304), Gemma 3 4B: 13.2 (#335)
| Benchmark | DeepSeek LLM 67B | Gemma 3 4B |
|---|---|---|
| Chess Puzzles | 0% | 0% |
| LMArena Hard Prompts | 1070 | 1253 |
| Epoch Capabilities Index | 110.5 | 116.02 |
| Kagi LLM Benchmark | — | 25.2% |
| DTBench | — | 50.9% |
| LMCA | — | 2.8% |
Math Gemma 3 4B leads
DeepSeek LLM 67B: 8.7 (#324), Gemma 3 4B: 16.8 (#292)
| Benchmark | DeepSeek LLM 67B | Gemma 3 4B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 7.5% |
| LMArena Math | 1108 | 1239 |
| MATH Level 5 | 6.4% | — |
Knowledge Gemma 3 4B leads
DeepSeek LLM 67B: 7.0 (#313), Gemma 3 4B: 11.8 (#299)
| Benchmark | DeepSeek LLM 67B | Gemma 3 4B |
|---|---|---|
| GPQA Diamond | 24.6% | 23.2% |
| Vectara Hallucination Rate | — | 6.4% |
| LMArena Expert | — | 1223 |
Multilingual Gemma 3 4B leads
DeepSeek LLM 67B: 29.4 (#267), Gemma 3 4B: 42.5 (#194)
| Benchmark | DeepSeek LLM 67B | Gemma 3 4B |
|---|---|---|
| LMArena Non-English | 1073 | 1273 |
| LMArena Chinese | 1132 | — |
| LMArena German | — | 1281 |
| LMArena Russian | — | 1294 |
Instruction Following Gemma 3 4B leads
DeepSeek LLM 67B: 55.4 (#277), Gemma 3 4B: 65.2 (#225)
| Benchmark | DeepSeek LLM 67B | Gemma 3 4B |
|---|---|---|
| LMArena Instruction Following | 1079 | 1239 |
Long Context Gemma 3 4B leads
DeepSeek LLM 67B: 33.1 (#265), Gemma 3 4B: 38.7 (#194)
| Benchmark | DeepSeek LLM 67B | Gemma 3 4B |
|---|---|---|
| LMArena Longer Query | 1092 | 1273 |
Writing & Preference Gemma 3 4B leads
DeepSeek LLM 67B: 31.6 (#282), Gemma 3 4B: 42.0 (#239)
| Benchmark | DeepSeek LLM 67B | Gemma 3 4B |
|---|---|---|
| LMArena Text | 1105 | 1291 |
| LMArena Creative Writing | 1067 | 1271 |
| LMArena Multi-Turn | 1082 | 1255 |
| EQ-Bench Creative Writing | — | 1068 |
Frequently asked questions
Is DeepSeek LLM 67B better than Gemma 3 4B?
Gemma 3 4B is the stronger model overall, scoring 28.1 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or Gemma 3 4B better for coding?
Gemma 3 4B scores higher on coding benchmarks: 35.9 versus 31.9 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Gemma 3 4B share?
13 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Gemma 3 4B has 22.