Model comparison
Qwen2.5-Coder-32B vs Tulu 3 (Tülu 3) 70B
Qwen2.5-Coder-32B and Tulu 3 (Tülu 3) 70B score almost the same on the Noometry Index (33.4 vs 33.0), so choose on price, context window or the category you care about most.
Last verified . 11 shared benchmarks.
Alibaba (Qwen)
Allen Institute for AI (Ai2)Summary
- They share 11 benchmarks with published results for both. Qwen2.5-Coder-32B scores higher in 3 categories and Tulu 3 (Tülu 3) 70B in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2.5-Coder-32B leads 33.3 to 14.2.
Side by side
| Qwen2.5-Coder-32B | Tulu 3 (Tülu 3) 70B | |
|---|---|---|
| Provider | Alibaba (Qwen) | Allen Institute for AI (Ai2) |
| Noometry Index | 33.4 | 33.0 |
| Released | 2024-09-18 | 2024-11-21 |
| Weights | Open | Open |
| Context window | 33K | — |
| Max output | 29K | — |
| Input $ / M tokens | $0.66 | — |
| Output $ / M tokens | $1 | — |
| Results tracked | 31 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Tulu 3 (Tülu 3) 70B leads
Qwen2.5-Coder-32B: 22.6 (#333), Tulu 3 (Tülu 3) 70B: 36.0 (#214)
| Benchmark | Qwen2.5-Coder-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Coding | 1276 | 1235 |
| SWE-bench Verified (bash only) | 9% | — |
| Aider Polyglot | 16.4% | — |
| BigCodeBench Instruct | 49% | — |
| LiveBench Coding | 56.9% | — |
| BigCodeBench Complete | 58% | — |
| HumanEval+ | 87.2% | — |
| MBPP+ | 77% | — |
Reasoning Tulu 3 (Tülu 3) 70B leads
Qwen2.5-Coder-32B: 21.2 (#225), Tulu 3 (Tülu 3) 70B: 23.9 (#169)
| Benchmark | Qwen2.5-Coder-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Hard Prompts | 1251 | 1220 |
| LiveBench Reasoning | 42.1% | — |
| LiveBench Data Analysis | 49.9% | — |
| Epoch Capabilities Index | 119.49 | — |
| HellaSwag | 83% | — |
| LiveBench | 46.2% | — |
| WinoGrande | 80.8% | — |
Math Qwen2.5-Coder-32B leads
Qwen2.5-Coder-32B: 33.3 (#204), Tulu 3 (Tülu 3) 70B: 14.2 (#303)
| Benchmark | Qwen2.5-Coder-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Math | 1251 | 1242 |
| OTIS Mock AIME 2024-2025 | — | 4.4% |
| LiveBench Math | 46.6% | — |
| MATH Level 5 | — | 42.7% |
| GSM8K | 93% | — |
Knowledge Qwen2.5-Coder-32B leads
Qwen2.5-Coder-32B: 33.4 (#203), Tulu 3 (Tülu 3) 70B: 25.0 (#264)
| Benchmark | Qwen2.5-Coder-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| GPQA Diamond | — | 46.3% |
| LMArena Expert | 1221 | — |
| ARC (AI2) Challenge | 70.5% | — |
| MMLU | 79.1% | — |
Multilingual Tulu 3 (Tülu 3) 70B leads
Qwen2.5-Coder-32B: 37.8 (#235), Tulu 3 (Tülu 3) 70B: 39.9 (#222)
| Benchmark | Qwen2.5-Coder-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Non-English | 1205 | 1236 |
| LMArena Chinese | 1222 | 1249 |
| LMArena Russian | 1228 | 1246 |
Instruction Following Tulu 3 (Tülu 3) 70B leads
Qwen2.5-Coder-32B: 61.4 (#245), Tulu 3 (Tülu 3) 70B: 64.8 (#227)
| Benchmark | Qwen2.5-Coder-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Instruction Following | 1223 | 1233 |
| LiveBench Instruction Following | 58.7% | — |
Long Context Too close to call
Qwen2.5-Coder-32B: 38.0 (#208), Tulu 3 (Tülu 3) 70B: 37.1 (#222)
| Benchmark | Qwen2.5-Coder-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Longer Query | 1251 | 1224 |
Writing & Preference Tulu 3 (Tülu 3) 70B leads
Qwen2.5-Coder-32B: 41.6 (#240), Tulu 3 (Tülu 3) 70B: 45.6 (#223)
| Benchmark | Qwen2.5-Coder-32B | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Text | 1230 | 1256 |
| LMArena Creative Writing | 1174 | 1231 |
| LMArena Multi-Turn | 1222 | 1252 |
| LiveBench Language | 23.3% | — |
Frequently asked questions
Is Qwen2.5-Coder-32B better than Tulu 3 (Tülu 3) 70B?
Qwen2.5-Coder-32B and Tulu 3 (Tülu 3) 70B score almost the same on the Noometry Index (33.4 vs 33.0), so choose on price, context window or the category you care about most.
Is Qwen2.5-Coder-32B or Tulu 3 (Tülu 3) 70B better for coding?
Tulu 3 (Tülu 3) 70B scores higher on coding benchmarks: 36.0 versus 22.6 in the Noometry coding category.
How many benchmarks do Qwen2.5-Coder-32B and Tulu 3 (Tülu 3) 70B share?
11 benchmarks have published results for both models. Qwen2.5-Coder-32B has 31 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.