Model comparison
Llama2 70b Steerlm Chat vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 32.9 to 31.2 on the Noometry Index.
Last verified . 9 shared benchmarks.
NVIDIA
Alibaba (Qwen)Summary
- They share 9 benchmarks with published results for both. Llama2 70b Steerlm Chat scores higher in 1 category and Qwen2.5-Coder-32B in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen2.5-Coder-32B leads 41.8 to 31.8.
Side by side
| Llama2 70b Steerlm Chat | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | NVIDIA | Alibaba (Qwen) |
| Noometry Index | 31.2 | 32.9 |
| Released | — | 2024-09-18 |
| Weights | Open | Open |
| Context window | — | 33K |
| Max output | — | 29K |
| Input $ / M tokens | — | $0.66 |
| Output $ / M tokens | — | $1 |
| Results tracked | 9 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama2 70b Steerlm Chat leads
Llama2 70b Steerlm Chat: 29.3 (#295), Qwen2.5-Coder-32B: 22.4 (#328)
| Benchmark | Llama2 70b Steerlm Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1025 | 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
Llama2 70b Steerlm Chat: 19.1 (#245), Qwen2.5-Coder-32B: 20.4 (#220)
| Benchmark | Llama2 70b Steerlm Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1047 | 1251 |
| LiveBench Reasoning | — | 42.1% |
| LiveBench Data Analysis | — | 49.9% |
| Epoch Capabilities Index | — | 119.41 |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math Qwen2.5-Coder-32B leads
Llama2 70b Steerlm Chat: 28.8 (#223), Qwen2.5-Coder-32B: 30.8 (#203)
| Benchmark | Llama2 70b Steerlm Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1072 | 1251 |
| LiveBench Math | — | 46.6% |
| GSM8K | — | 93% |
Knowledge Not comparable
Llama2 70b Steerlm Chat: —, Qwen2.5-Coder-32B: 32.9 (#200)
| Benchmark | Llama2 70b Steerlm Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | — | 1221 |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multilingual Qwen2.5-Coder-32B leads
Llama2 70b Steerlm Chat: 28.9 (#265), Qwen2.5-Coder-32B: 38.0 (#230)
| Benchmark | Llama2 70b Steerlm Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1064 | 1205 |
| LMArena Chinese | — | 1222 |
| LMArena Russian | — | 1228 |
Instruction Following Qwen2.5-Coder-32B leads
Llama2 70b Steerlm Chat: 54.5 (#274), Qwen2.5-Coder-32B: 61.6 (#240)
| Benchmark | Llama2 70b Steerlm Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1061 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Qwen2.5-Coder-32B leads
Llama2 70b Steerlm Chat: 30.4 (#283), Qwen2.5-Coder-32B: 37.9 (#204)
| Benchmark | Llama2 70b Steerlm Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 998 | 1251 |
Writing & Preference Qwen2.5-Coder-32B leads
Llama2 70b Steerlm Chat: 31.8 (#278), Qwen2.5-Coder-32B: 41.8 (#235)
| Benchmark | Llama2 70b Steerlm Chat | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1098 | 1230 |
| LMArena Creative Writing | 1091 | 1174 |
| LMArena Multi-Turn | 1058 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Llama2 70b Steerlm Chat better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 32.9 to 31.2 on the Noometry Index.
Is Llama2 70b Steerlm Chat or Qwen2.5-Coder-32B better for coding?
Llama2 70b Steerlm Chat scores higher on coding benchmarks: 29.3 versus 22.4 in the Noometry coding category.
How many benchmarks do Llama2 70b Steerlm Chat and Qwen2.5-Coder-32B share?
9 benchmarks have published results for both models. Llama2 70b Steerlm Chat has 9 scored results on Noometry and Qwen2.5-Coder-32B has 31.