Model comparison
Llama 3-8B vs Tulu 3 (Tülu 3) 70B
Tulu 3 (Tülu 3) 70B is the stronger model overall, scoring 33.0 to 25.5 on the Noometry Index.
Last verified . 14 shared benchmarks.
Meta
Allen Institute for AI (Ai2)Summary
- They share 14 benchmarks with published results for both. Llama 3-8B scores higher in 0 categories and Tulu 3 (Tülu 3) 70B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Tulu 3 (Tülu 3) 70B leads 25.0 to 7.8.
- The biggest single-benchmark swing is MATH Level 5: 6.1% for Llama 3-8B and 42.7% for Tulu 3 (Tülu 3) 70B.
Side by side
| Llama 3-8B | Tulu 3 (Tülu 3) 70B | |
|---|---|---|
| Provider | Meta | Allen Institute for AI (Ai2) |
| Noometry Index | 25.5 | 33.0 |
| Released | 2024-04-18 | 2024-11-21 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| 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
Llama 3-8B: 31.0 (#289), Tulu 3 (Tülu 3) 70B: 36.0 (#214)
| Benchmark | Llama 3-8B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Coding | 1152 | 1235 |
| BigCodeBench Instruct | 31.9% | — |
| BigCodeBench Complete | 36.9% | — |
| HumanEval+ | 56.7% | — |
| MBPP+ | 54.8% | — |
Reasoning Tulu 3 (Tülu 3) 70B leads
Llama 3-8B: 14.3 (#326), Tulu 3 (Tülu 3) 70B: 23.9 (#169)
| Benchmark | Llama 3-8B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Hard Prompts | 1133 | 1220 |
| Chess Puzzles | 0% | — |
| DTBench | 43.9% | — |
| Adversarial NLI | 57.3% | — |
| Epoch Capabilities Index | 116.45 | — |
| ForecastBench | 58.6 | — |
| WinoGrande | 75.7% | — |
Math Tulu 3 (Tülu 3) 70B leads
Llama 3-8B: 8.8 (#323), Tulu 3 (Tülu 3) 70B: 14.2 (#303)
| Benchmark | Llama 3-8B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.9% | 4.4% |
| LMArena Math | 1151 | 1242 |
| MATH Level 5 | 6.1% | 42.7% |
Knowledge Tulu 3 (Tülu 3) 70B leads
Llama 3-8B: 7.8 (#308), Tulu 3 (Tülu 3) 70B: 25.0 (#264)
| Benchmark | Llama 3-8B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| GPQA Diamond | 26.1% | 46.3% |
| LMArena Expert | 1113 | — |
| ARC (AI2) Challenge | 82.8% | — |
| MMLU | 68.8% | — |
| OpenBookQA | 82.6% | — |
| TriviaQA | 67.7% | — |
Multilingual Tulu 3 (Tülu 3) 70B leads
Llama 3-8B: 30.8 (#261), Tulu 3 (Tülu 3) 70B: 39.9 (#222)
| Benchmark | Llama 3-8B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Non-English | 1098 | 1236 |
| LMArena Chinese | 1076 | 1249 |
| LMArena Russian | 1109 | 1246 |
| LMArena French | 1159 | — |
| LMArena German | 1104 | — |
| LMArena Japanese | 967 | — |
| LMArena Korean | 1004 | — |
| LMArena Spanish | 1173 | — |
Instruction Following Tulu 3 (Tülu 3) 70B leads
Llama 3-8B: 58.4 (#260), Tulu 3 (Tülu 3) 70B: 64.8 (#227)
| Benchmark | Llama 3-8B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Instruction Following | 1127 | 1233 |
Long Context Tulu 3 (Tülu 3) 70B leads
Llama 3-8B: 34.2 (#251), Tulu 3 (Tülu 3) 70B: 37.1 (#222)
| Benchmark | Llama 3-8B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Longer Query | 1128 | 1224 |
Writing & Preference Tulu 3 (Tülu 3) 70B leads
Llama 3-8B: 37.5 (#256), Tulu 3 (Tülu 3) 70B: 45.6 (#223)
| Benchmark | Llama 3-8B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Text | 1166 | 1256 |
| LMArena Creative Writing | 1150 | 1231 |
| LMArena Multi-Turn | 1152 | 1252 |
Frequently asked questions
Is Llama 3-8B better than Tulu 3 (Tülu 3) 70B?
Tulu 3 (Tülu 3) 70B is the stronger model overall, scoring 33.0 to 25.5 on the Noometry Index.
Is Llama 3-8B or Tulu 3 (Tülu 3) 70B better for coding?
Tulu 3 (Tülu 3) 70B scores higher on coding benchmarks: 36.0 versus 31.0 in the Noometry coding category.
How many benchmarks do Llama 3-8B and Tulu 3 (Tülu 3) 70B share?
14 benchmarks have published results for both models. Llama 3-8B has 34 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.