Model comparison
Gemma 2 9B vs Tulu 3 (Tülu 3) 70B
Tulu 3 (Tülu 3) 70B is the stronger model overall, scoring 33.0 to 25.9 on the Noometry Index.
Last verified . 14 shared benchmarks.
Google
Allen Institute for AI (Ai2)Summary
- They share 14 benchmarks with published results for both. Gemma 2 9B scores higher in 0 categories and Tulu 3 (Tülu 3) 70B in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Tulu 3 (Tülu 3) 70B leads 25.0 to 9.7.
- The biggest single-benchmark swing is MATH Level 5: 21% for Gemma 2 9B and 42.7% for Tulu 3 (Tülu 3) 70B.
Side by side
| Gemma 2 9B | Tulu 3 (Tülu 3) 70B | |
|---|---|---|
| Provider | Allen Institute for AI (Ai2) | |
| Noometry Index | 25.9 | 33.0 |
| Released | 2024-06-24 | 2024-11-21 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 35 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Tulu 3 (Tülu 3) 70B leads
Gemma 2 9B: 29.4 (#304), Tulu 3 (Tülu 3) 70B: 36.0 (#214)
| Benchmark | Gemma 2 9B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Coding | 1173 | 1235 |
| BigCodeBench Instruct | 34.7% | — |
| LiveBench Coding | 22.5% | — |
| BigCodeBench Complete | 40.6% | — |
Reasoning Tulu 3 (Tülu 3) 70B leads
Gemma 2 9B: 15.9 (#309), Tulu 3 (Tülu 3) 70B: 23.9 (#169)
| Benchmark | Gemma 2 9B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Hard Prompts | 1171 | 1220 |
| LiveBench Reasoning | 15.2% | — |
| LiveBench Data Analysis | 36.4% | — |
| Epoch Capabilities Index | 119.83 | — |
| LiveBench | 28.7% | — |
| PIQA | 83.7% | — |
Math Tulu 3 (Tülu 3) 70B leads
Gemma 2 9B: 9.9 (#318), Tulu 3 (Tülu 3) 70B: 14.2 (#303)
| Benchmark | Gemma 2 9B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.6% | 4.4% |
| LMArena Math | 1183 | 1242 |
| MATH Level 5 | 21% | 42.7% |
| LiveBench Math | 19.8% | — |
| GSM8K | 84.9% | — |
Knowledge Tulu 3 (Tülu 3) 70B leads
Gemma 2 9B: 9.7 (#305), Tulu 3 (Tülu 3) 70B: 25.0 (#264)
| Benchmark | Gemma 2 9B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| GPQA Diamond | 27.5% | 46.3% |
| LMArena Expert | 1147 | — |
| BoolQ | 85.7% | — |
| MMLU | 72.1% | — |
Multilingual Tulu 3 (Tülu 3) 70B leads
Gemma 2 9B: 36.6 (#238), Tulu 3 (Tülu 3) 70B: 39.9 (#222)
| Benchmark | Gemma 2 9B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Non-English | 1188 | 1236 |
| LMArena Chinese | 1185 | 1249 |
| LMArena Russian | 1200 | 1246 |
| LMArena French | 1190 | — |
| LMArena German | 1186 | — |
| LMArena Japanese | 1144 | — |
| LMArena Korean | 1137 | — |
| LMArena Spanish | 1200 | — |
Instruction Following Tulu 3 (Tülu 3) 70B leads
Gemma 2 9B: 57.6 (#269), Tulu 3 (Tülu 3) 70B: 64.8 (#227)
| Benchmark | Gemma 2 9B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Instruction Following | 1178 | 1233 |
| LiveBench Instruction Following | 52.6% | — |
Long Context Too close to call
Gemma 2 9B: 36.3 (#233), Tulu 3 (Tülu 3) 70B: 37.1 (#222)
| Benchmark | Gemma 2 9B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Longer Query | 1197 | 1224 |
Writing & Preference Tulu 3 (Tülu 3) 70B leads
Gemma 2 9B: 32.1 (#281), Tulu 3 (Tülu 3) 70B: 45.6 (#223)
| Benchmark | Gemma 2 9B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Text | 1207 | 1256 |
| LMArena Creative Writing | 1206 | 1231 |
| LMArena Multi-Turn | 1193 | 1252 |
| EQ-Bench Creative Writing | 841 | — |
| LiveBench Language | 25.5% | — |
Frequently asked questions
Is Gemma 2 9B better than Tulu 3 (Tülu 3) 70B?
Tulu 3 (Tülu 3) 70B is the stronger model overall, scoring 33.0 to 25.9 on the Noometry Index.
Is Gemma 2 9B or Tulu 3 (Tülu 3) 70B better for coding?
Tulu 3 (Tülu 3) 70B scores higher on coding benchmarks: 36.0 versus 29.4 in the Noometry coding category.
How many benchmarks do Gemma 2 9B and Tulu 3 (Tülu 3) 70B share?
14 benchmarks have published results for both models. Gemma 2 9B has 35 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.