Model comparison
Longcat Flash Chat vs Qwen2.5-Coder-32B
Longcat Flash Chat is the stronger model overall, scoring 42.1 to 33.4 on the Noometry Index.
Last verified . 12 shared benchmarks.
Alibaba (Qwen)Summary
- They share 12 benchmarks with published results for both. Longcat Flash Chat scores higher in 7 categories and Qwen2.5-Coder-32B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where Longcat Flash Chat leads 43.5 to 22.6.
Side by side
| Longcat Flash Chat | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Meituan | Alibaba (Qwen) |
| Noometry Index | 42.1 | 33.4 |
| Released | — | 2024-09-18 |
| Weights | Open | Open |
| Context window | — | 33K |
| Max output | — | 29K |
| Input $ / M tokens | — | $0.66 |
| Output $ / M tokens | — | $1 |
| Results tracked | 19 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Longcat Flash Chat leads
Longcat Flash Chat: 43.5 (#87), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Longcat Flash Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1471 | 1276 |
| 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 Qwen2.5-Coder-32B leads
Longcat Flash Chat: 19.0 (#272), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Longcat Flash Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1440 | 1251 |
| Kagi LLM Benchmark | 43.9% | — |
| NYT Connections (extended) | 17.7% | — |
| LiveBench Reasoning | — | 42.1% |
| LiveBench Data Analysis | — | 49.9% |
| Epoch Capabilities Index | — | 119.49 |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math Longcat Flash Chat leads
Longcat Flash Chat: 39.4 (#107), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Longcat Flash Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1442 | 1251 |
| LiveBench Math | — | 46.6% |
| GSM8K | — | 93% |
Knowledge Longcat Flash Chat leads
Longcat Flash Chat: 40.6 (#116), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Longcat Flash Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1454 | 1221 |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multilingual Longcat Flash Chat leads
Longcat Flash Chat: 51.9 (#101), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Longcat Flash Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1404 | 1205 |
| LMArena Chinese | 1465 | 1222 |
| LMArena Russian | 1395 | 1228 |
| LMArena French | 1456 | — |
| LMArena German | 1408 | — |
| LMArena Japanese | 1373 | — |
| LMArena Korean | 1371 | — |
| LMArena Spanish | 1445 | — |
Instruction Following Longcat Flash Chat leads
Longcat Flash Chat: 74.4 (#96), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Longcat Flash Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1411 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Longcat Flash Chat leads
Longcat Flash Chat: 43.5 (#93), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Longcat Flash Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1425 | 1251 |
Writing & Preference Longcat Flash Chat leads
Longcat Flash Chat: 61.0 (#91), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Longcat Flash Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1427 | 1230 |
| LMArena Creative Writing | 1388 | 1174 |
| LMArena Multi-Turn | 1418 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Longcat Flash Chat better than Qwen2.5-Coder-32B?
Longcat Flash Chat is the stronger model overall, scoring 42.1 to 33.4 on the Noometry Index.
Is Longcat Flash Chat or Qwen2.5-Coder-32B better for coding?
Longcat Flash Chat scores higher on coding benchmarks: 43.5 versus 22.6 in the Noometry coding category.
How many benchmarks do Longcat Flash Chat and Qwen2.5-Coder-32B share?
12 benchmarks have published results for both models. Longcat Flash Chat has 19 scored results on Noometry and Qwen2.5-Coder-32B has 31.