Model comparison
DeepSeek LLM 67B vs Gemma 2B
Gemma 2B is the stronger model overall, scoring 29.6 to 24.9 on the Noometry Index.
Last verified . 11 shared benchmarks.
DeepSeek
GoogleSummary
- They share 11 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 5 categories and Gemma 2B in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 2B leads 30.0 to 8.7.
Side by side
| DeepSeek LLM 67B | Gemma 2B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 24.9 | 29.6 |
| Released | 2023-11-29 | 2024-02-21 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 15 | 23 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek LLM 67B leads
DeepSeek LLM 67B: 31.9 (#278), Gemma 2B: 29.4 (#305)
| Benchmark | DeepSeek LLM 67B | Gemma 2B |
|---|---|---|
| LMArena Coding | 1096 | 1010 |
| HumanEval+ | — | 20.7% |
| MBPP+ | — | 34.1% |
Reasoning Gemma 2B leads
DeepSeek LLM 67B: 16.5 (#304), Gemma 2B: 18.8 (#275)
| Benchmark | DeepSeek LLM 67B | Gemma 2B |
|---|---|---|
| LMArena Hard Prompts | 1070 | 989 |
| Epoch Capabilities Index | 110.5 | 94.2 |
| Chess Puzzles | 0% | — |
| BIG-Bench Hard | — | 35.2% |
| HellaSwag | — | 71.4% |
| PIQA | — | 77.3% |
| WinoGrande | — | 65.4% |
Math Gemma 2B leads
DeepSeek LLM 67B: 8.7 (#324), Gemma 2B: 30.0 (#239)
| Benchmark | DeepSeek LLM 67B | Gemma 2B |
|---|---|---|
| LMArena Math | 1108 | 1009 |
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| MATH Level 5 | 6.4% | — |
| GSM8K | — | 17.7% |
Knowledge Not comparable
DeepSeek LLM 67B: 7.0 (#313), Gemma 2B: —
| Benchmark | DeepSeek LLM 67B | Gemma 2B |
|---|---|---|
| GPQA Diamond | 24.6% | — |
| ARC (AI2) Challenge | — | 42.1% |
| BoolQ | — | 69.4% |
| MMLU | — | 42.3% |
| TriviaQA | — | 53.2% |
Multilingual DeepSeek LLM 67B leads
DeepSeek LLM 67B: 29.4 (#267), Gemma 2B: 23.0 (#294)
| Benchmark | DeepSeek LLM 67B | Gemma 2B |
|---|---|---|
| LMArena Non-English | 1073 | 958 |
| LMArena Chinese | 1132 | 986 |
| LMArena Russian | — | 937 |
Instruction Following DeepSeek LLM 67B leads
DeepSeek LLM 67B: 55.4 (#277), Gemma 2B: 48.5 (#302)
| Benchmark | DeepSeek LLM 67B | Gemma 2B |
|---|---|---|
| LMArena Instruction Following | 1079 | 970 |
Long Context DeepSeek LLM 67B leads
DeepSeek LLM 67B: 33.1 (#265), Gemma 2B: 29.9 (#291)
| Benchmark | DeepSeek LLM 67B | Gemma 2B |
|---|---|---|
| LMArena Longer Query | 1092 | 981 |
Writing & Preference DeepSeek LLM 67B leads
DeepSeek LLM 67B: 31.6 (#282), Gemma 2B: 24.0 (#308)
| Benchmark | DeepSeek LLM 67B | Gemma 2B |
|---|---|---|
| LMArena Text | 1105 | 1002 |
| LMArena Creative Writing | 1067 | 987 |
| LMArena Multi-Turn | 1082 | 945 |
Frequently asked questions
Is DeepSeek LLM 67B better than Gemma 2B?
Gemma 2B is the stronger model overall, scoring 29.6 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or Gemma 2B better for coding?
DeepSeek LLM 67B scores higher on coding benchmarks: 31.9 versus 29.4 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Gemma 2B share?
11 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Gemma 2B has 23.