Model comparison
Gemma 2 27B vs Tulu 3 (Tülu 3) 70B
Tulu 3 (Tülu 3) 70B is the stronger model overall, scoring 33.0 to 29.4 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 27B scores higher in 1 category and Tulu 3 (Tülu 3) 70B in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Tulu 3 (Tülu 3) 70B leads 23.9 to 15.3.
- The biggest single-benchmark swing is MATH Level 5: 27.9% for Gemma 2 27B and 42.7% for Tulu 3 (Tülu 3) 70B.
Side by side
| Gemma 2 27B | Tulu 3 (Tülu 3) 70B | |
|---|---|---|
| Provider | Allen Institute for AI (Ai2) | |
| Noometry Index | 29.4 | 33.0 |
| Released | 2024-06-24 | 2024-11-21 |
| Weights | Open | Open |
| Context window | 8K | — |
| Max output | 2K | — |
| Input $ / M tokens | $0.65 | — |
| Output $ / M tokens | $0.65 | — |
| Results tracked | 34 | 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 27B: 34.1 (#246), Tulu 3 (Tülu 3) 70B: 36.0 (#214)
| Benchmark | Gemma 2 27B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Coding | 1211 | 1235 |
| BigCodeBench Instruct | 42.8% | — |
| LiveBench Coding | 36% | — |
| BigCodeBench Complete | 52.5% | — |
Reasoning Tulu 3 (Tülu 3) 70B leads
Gemma 2 27B: 15.3 (#315), Tulu 3 (Tülu 3) 70B: 23.9 (#169)
| Benchmark | Gemma 2 27B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Hard Prompts | 1198 | 1220 |
| LiveBench Reasoning | 28.1% | — |
| DTBench | 48% | — |
| LiveBench Data Analysis | 47.9% | — |
| LMCA | 7.1% | — |
| Epoch Capabilities Index | 122.08 | — |
| LiveBench | 38.2% | — |
Math Tulu 3 (Tülu 3) 70B leads
Gemma 2 27B: 10.7 (#311), Tulu 3 (Tülu 3) 70B: 14.2 (#303)
| Benchmark | Gemma 2 27B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.4% | 4.4% |
| LMArena Math | 1212 | 1242 |
| MATH Level 5 | 27.9% | 42.7% |
| LiveBench Math | 26.5% | — |
Knowledge Tulu 3 (Tülu 3) 70B leads
Gemma 2 27B: 19.0 (#280), Tulu 3 (Tülu 3) 70B: 25.0 (#264)
| Benchmark | Gemma 2 27B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| GPQA Diamond | 36.5% | 46.3% |
| Confabulations | 27.1% | — |
| LMArena Expert | 1172 | — |
| MMLU | 75.7% | — |
Multilingual Tulu 3 (Tülu 3) 70B leads
Gemma 2 27B: 38.6 (#226), Tulu 3 (Tülu 3) 70B: 39.9 (#222)
| Benchmark | Gemma 2 27B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Non-English | 1217 | 1236 |
| LMArena Chinese | 1221 | 1249 |
| LMArena Russian | 1234 | 1246 |
| LMArena French | 1247 | — |
| LMArena German | 1209 | — |
| LMArena Japanese | 1175 | — |
| LMArena Korean | 1174 | — |
| LMArena Spanish | 1228 | — |
Instruction Following Tulu 3 (Tülu 3) 70B leads
Gemma 2 27B: 60.5 (#249), Tulu 3 (Tülu 3) 70B: 64.8 (#227)
| Benchmark | Gemma 2 27B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Instruction Following | 1206 | 1233 |
| LiveBench Instruction Following | 58.1% | — |
Long Context Too close to call
Gemma 2 27B: 37.3 (#218), Tulu 3 (Tülu 3) 70B: 37.1 (#222)
| Benchmark | Gemma 2 27B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Longer Query | 1231 | 1224 |
Writing & Preference Tulu 3 (Tülu 3) 70B leads
Gemma 2 27B: 44.2 (#225), Tulu 3 (Tülu 3) 70B: 45.6 (#223)
| Benchmark | Gemma 2 27B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Text | 1231 | 1256 |
| LMArena Creative Writing | 1241 | 1231 |
| LMArena Multi-Turn | 1224 | 1252 |
| LiveBench Language | 32.6% | — |
Frequently asked questions
Is Gemma 2 27B better than Tulu 3 (Tülu 3) 70B?
Tulu 3 (Tülu 3) 70B is the stronger model overall, scoring 33.0 to 29.4 on the Noometry Index.
Is Gemma 2 27B or Tulu 3 (Tülu 3) 70B better for coding?
Tulu 3 (Tülu 3) 70B scores higher on coding benchmarks: 36.0 versus 34.1 in the Noometry coding category.
How many benchmarks do Gemma 2 27B and Tulu 3 (Tülu 3) 70B share?
14 benchmarks have published results for both models. Gemma 2 27B has 34 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.