Model comparison
phi-3-medium 14B vs Tulu 3 (Tülu 3) 70B
Tulu 3 (Tülu 3) 70B is the stronger model overall, scoring 33.0 to 29.7 on the Noometry Index.
Last verified . 2 shared benchmarks.
Microsoft
Allen Institute for AI (Ai2)Summary
- They share 2 benchmarks with published results for both. phi-3-medium 14B scores higher in 2 categories and Tulu 3 (Tülu 3) 70B in 1 category; 2 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Tulu 3 (Tülu 3) 70B leads 25.0 to 9.1.
- The biggest single-benchmark swing is MATH Level 5: 17.6% for phi-3-medium 14B and 42.7% for Tulu 3 (Tülu 3) 70B.
Side by side
| phi-3-medium 14B | Tulu 3 (Tülu 3) 70B | |
|---|---|---|
| Provider | Microsoft | Allen Institute for AI (Ai2) |
| Noometry Index | 29.7 | 33.0 |
| Released | 2024-04-23 | 2024-11-21 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 13 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
phi-3-medium 14B: 36.8 (#201), Tulu 3 (Tülu 3) 70B: 36.0 (#214)
| Benchmark | phi-3-medium 14B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| BigCodeBench Instruct | 37.6% | — |
| LMArena Coding | — | 1235 |
| BigCodeBench Complete | 48.7% | — |
Reasoning Not comparable
phi-3-medium 14B: —, Tulu 3 (Tülu 3) 70B: 23.9 (#169)
| Benchmark | phi-3-medium 14B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Hard Prompts | — | 1220 |
| Adversarial NLI | 55.8% | — |
| BIG-Bench Hard | 81.4% | — |
| Epoch Capabilities Index | 121.23 | — |
| HellaSwag | 82.4% | — |
| WinoGrande | 81.5% | — |
Math phi-3-medium 14B leads
phi-3-medium 14B: 27.3 (#250), Tulu 3 (Tülu 3) 70B: 14.2 (#303)
| Benchmark | phi-3-medium 14B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| MATH Level 5 | 17.6% | 42.7% |
| OTIS Mock AIME 2024-2025 | — | 4.4% |
| LMArena Math | — | 1242 |
Knowledge Tulu 3 (Tülu 3) 70B leads
phi-3-medium 14B: 9.1 (#306), Tulu 3 (Tülu 3) 70B: 25.0 (#264)
| Benchmark | phi-3-medium 14B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| GPQA Diamond | 27.6% | 46.3% |
| ARC (AI2) Challenge | 91.6% | — |
| MMLU | 78% | — |
| OpenBookQA | 87.4% | — |
| TriviaQA | 73.9% | — |
Multilingual Not comparable
phi-3-medium 14B: —, Tulu 3 (Tülu 3) 70B: 39.9 (#222)
| Benchmark | phi-3-medium 14B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Non-English | — | 1236 |
| LMArena Chinese | — | 1249 |
| LMArena Russian | — | 1246 |
Instruction Following Not comparable
phi-3-medium 14B: —, Tulu 3 (Tülu 3) 70B: 64.8 (#227)
| Benchmark | phi-3-medium 14B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Instruction Following | — | 1233 |
Long Context Not comparable
phi-3-medium 14B: —, Tulu 3 (Tülu 3) 70B: 37.1 (#222)
| Benchmark | phi-3-medium 14B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Longer Query | — | 1224 |
Writing & Preference Not comparable
phi-3-medium 14B: —, Tulu 3 (Tülu 3) 70B: 45.6 (#223)
| Benchmark | phi-3-medium 14B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Text | — | 1256 |
| LMArena Creative Writing | — | 1231 |
| LMArena Multi-Turn | — | 1252 |
Frequently asked questions
Is phi-3-medium 14B better than Tulu 3 (Tülu 3) 70B?
Tulu 3 (Tülu 3) 70B is the stronger model overall, scoring 33.0 to 29.7 on the Noometry Index.
Is phi-3-medium 14B or Tulu 3 (Tülu 3) 70B better for coding?
They score almost the same on coding (36.8 vs 36.0); test both on your own repository before choosing.
How many benchmarks do phi-3-medium 14B and Tulu 3 (Tülu 3) 70B share?
2 benchmarks have published results for both models. phi-3-medium 14B has 13 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.