Model comparison
Qwen1.5-32B vs Tulu 3 (Tülu 3) 70B
Tulu 3 (Tülu 3) 70B is the stronger model overall, scoring 33.0 to 30.5 on the Noometry Index.
Last verified . 12 shared benchmarks.
Alibaba (Qwen)
Allen Institute for AI (Ai2)Summary
- They share 12 benchmarks with published results for both. Qwen1.5-32B scores higher in 1 category and Tulu 3 (Tülu 3) 70B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen1.5-32B leads 33.0 to 14.2.
- The biggest single-benchmark swing is GPQA Diamond: 30.7% for Qwen1.5-32B and 46.3% for Tulu 3 (Tülu 3) 70B.
Side by side
| Qwen1.5-32B | Tulu 3 (Tülu 3) 70B | |
|---|---|---|
| Provider | Alibaba (Qwen) | Allen Institute for AI (Ai2) |
| Noometry Index | 30.5 | 33.0 |
| Released | 2024-02-04 | 2024-11-21 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 21 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Tulu 3 (Tülu 3) 70B leads
Qwen1.5-32B: 31.7 (#282), Tulu 3 (Tülu 3) 70B: 36.0 (#214)
| Benchmark | Qwen1.5-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Coding | 1155 | 1235 |
| BigCodeBench Instruct | 32.3% | — |
| BigCodeBench Complete | 42% | — |
Reasoning Tulu 3 (Tülu 3) 70B leads
Qwen1.5-32B: 21.8 (#212), Tulu 3 (Tülu 3) 70B: 23.9 (#169)
| Benchmark | Qwen1.5-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Hard Prompts | 1130 | 1220 |
Math Qwen1.5-32B leads
Qwen1.5-32B: 33.0 (#207), Tulu 3 (Tülu 3) 70B: 14.2 (#303)
| Benchmark | Qwen1.5-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Math | 1155 | 1242 |
| OTIS Mock AIME 2024-2025 | — | 4.4% |
| MATH Level 5 | — | 42.7% |
Knowledge Tulu 3 (Tülu 3) 70B leads
Qwen1.5-32B: 13.5 (#296), Tulu 3 (Tülu 3) 70B: 25.0 (#264)
| Benchmark | Qwen1.5-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| GPQA Diamond | 30.7% | 46.3% |
| LMArena Expert | 1126 | — |
| MMLU | 74.4% | — |
Multilingual Tulu 3 (Tülu 3) 70B leads
Qwen1.5-32B: 31.4 (#259), Tulu 3 (Tülu 3) 70B: 39.9 (#222)
| Benchmark | Qwen1.5-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Non-English | 1106 | 1236 |
| LMArena Chinese | 1177 | 1249 |
| LMArena Russian | 1073 | 1246 |
| LMArena French | 1101 | — |
| LMArena German | 1058 | — |
| LMArena Japanese | 1027 | — |
| LMArena Korean | 1008 | — |
| LMArena Spanish | 1089 | — |
Instruction Following Tulu 3 (Tülu 3) 70B leads
Qwen1.5-32B: 57.7 (#265), Tulu 3 (Tülu 3) 70B: 64.8 (#227)
| Benchmark | Qwen1.5-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Instruction Following | 1116 | 1233 |
Long Context Tulu 3 (Tülu 3) 70B leads
Qwen1.5-32B: 34.7 (#246), Tulu 3 (Tülu 3) 70B: 37.1 (#222)
| Benchmark | Qwen1.5-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Longer Query | 1146 | 1224 |
Writing & Preference Tulu 3 (Tülu 3) 70B leads
Qwen1.5-32B: 34.2 (#271), Tulu 3 (Tülu 3) 70B: 45.6 (#223)
| Benchmark | Qwen1.5-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Text | 1137 | 1256 |
| LMArena Creative Writing | 1083 | 1231 |
| LMArena Multi-Turn | 1140 | 1252 |
Frequently asked questions
Is Qwen1.5-32B better than Tulu 3 (Tülu 3) 70B?
Tulu 3 (Tülu 3) 70B is the stronger model overall, scoring 33.0 to 30.5 on the Noometry Index.
Is Qwen1.5-32B or Tulu 3 (Tülu 3) 70B better for coding?
Tulu 3 (Tülu 3) 70B scores higher on coding benchmarks: 36.0 versus 31.7 in the Noometry coding category.
How many benchmarks do Qwen1.5-32B and Tulu 3 (Tülu 3) 70B share?
12 benchmarks have published results for both models. Qwen1.5-32B has 21 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.