Model comparison
Llama-3.3-70B-Instruct vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 32.9 to 30.3 on the Noometry Index. Llama-3.3-70B-Instruct costs 4.8× less per token, which makes it the better buy when Qwen2.5-Coder-32B's lead doesn't matter for your workload.
Last verified . 23 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. Llama-3.3-70B-Instruct scores higher in 4 categories and Qwen2.5-Coder-32B in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2.5-Coder-32B leads 30.8 to 14.3.
- The biggest single-benchmark swing is LiveBench Instruction Following: 82.7% for Llama-3.3-70B-Instruct and 58.7% for Qwen2.5-Coder-32B.
- Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- Llama-3.3-70B-Instruct accepts more context: 128K tokens versus 33K.
Side by side
| Llama-3.3-70B-Instruct | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 30.3 | 32.9 |
| Released | 2024-12-06 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 128K | 33K |
| Max output | 4K | 29K |
| Input $ / M tokens | $0.10 | $0.66 |
| Output $ / M tokens | $0.32 | $1 |
| Results tracked | 43 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 30.6 (#284), Qwen2.5-Coder-32B: 22.4 (#328)
| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| BigCodeBench Instruct | 46.9% | 49% |
| LiveBench Coding | 36.6% | 56.9% |
| LMArena Coding | 1268 | 1276 |
| BigCodeBench Complete | 57.5% | 58% |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| SciCode | 26% | — |
| WeirdML | 14.4% | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Agentic & Tool Use Not comparable
Llama-3.3-70B-Instruct: 25.6 (#100), Qwen2.5-Coder-32B: —
| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 31.9% | — |
| BALROG | 23% | — |
Reasoning Qwen2.5-Coder-32B leads
Llama-3.3-70B-Instruct: 13.7 (#325), Qwen2.5-Coder-32B: 20.4 (#220)
| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Reasoning | 50.8% | 42.1% |
| LMArena Hard Prompts | 1257 | 1251 |
| LiveBench Data Analysis | 49.5% | 49.9% |
| Epoch Capabilities Index | 127.32 | 119.41 |
| LiveBench | 50.2% | 46.2% |
| SimpleBench | 19.9% | — |
| CritPt | 0% | — |
| DTBench | 59.5% | — |
| LMCA | 17.5% | — |
| ForecastBench | 58.6 | — |
| HellaSwag | — | 83% |
| WinoGrande | — | 80.8% |
Math Qwen2.5-Coder-32B leads
Llama-3.3-70B-Instruct: 14.3 (#292), Qwen2.5-Coder-32B: 30.8 (#203)
| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Math | 42.2% | 46.6% |
| LMArena Math | 1268 | 1251 |
| OTIS Mock AIME 2024-2025 | 5.1% | — |
| MATH Level 5 | 41.6% | — |
| GSM8K | — | 93% |
Knowledge Qwen2.5-Coder-32B leads
Llama-3.3-70B-Instruct: 30.5 (#220), Qwen2.5-Coder-32B: 32.9 (#200)
| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1225 | 1221 |
| MMLU | 86.3% | 79.1% |
| GPQA Diamond | 47.4% | — |
| Confabulations | 22.8% | — |
| Vectara Hallucination Rate | 4.1% | — |
| ARC (AI2) Challenge | — | 70.5% |
Multilingual Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 40.1 (#215), Qwen2.5-Coder-32B: 38.0 (#230)
| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1236 | 1205 |
| LMArena Chinese | 1217 | 1222 |
| LMArena Russian | 1252 | 1228 |
| LMArena French | 1280 | — |
| LMArena German | 1251 | — |
| LMArena Japanese | 1149 | — |
| LMArena Korean | 1142 | — |
| LMArena Spanish | 1269 | — |
Instruction Following Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 71.3 (#152), Qwen2.5-Coder-32B: 61.6 (#240)
| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Instruction Following | 82.7% | 58.7% |
| LMArena Instruction Following | 1242 | 1223 |
Long Context Qwen2.5-Coder-32B leads
Llama-3.3-70B-Instruct: 26.4 (#290), Qwen2.5-Coder-32B: 37.9 (#204)
| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1256 | 1251 |
| Fiction.LiveBench | 33.3% | — |
Writing & Preference Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 47.8 (#203), Qwen2.5-Coder-32B: 41.8 (#235)
| Benchmark | Llama-3.3-70B-Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1274 | 1230 |
| LMArena Creative Writing | 1250 | 1174 |
| LMArena Multi-Turn | 1279 | 1222 |
| LiveBench Language | 39.2% | 23.3% |
Frequently asked questions
Is Llama-3.3-70B-Instruct better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 32.9 to 30.3 on the Noometry Index. Llama-3.3-70B-Instruct costs 4.8× less per token, which makes it the better buy when Qwen2.5-Coder-32B's lead doesn't matter for your workload.
Which is cheaper, Llama-3.3-70B-Instruct or Qwen2.5-Coder-32B?
Llama-3.3-70B-Instruct is cheaper. It lists at $0.10 per million input tokens and $0.32 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is Llama-3.3-70B-Instruct or Qwen2.5-Coder-32B better for coding?
Llama-3.3-70B-Instruct scores higher on coding benchmarks: 30.6 versus 22.4 in the Noometry coding category.
Which has the bigger context window?
Llama-3.3-70B-Instruct does, with 128K tokens against 33K.
How many benchmarks do Llama-3.3-70B-Instruct and Qwen2.5-Coder-32B share?
23 benchmarks have published results for both models. Llama-3.3-70B-Instruct has 43 scored results on Noometry and Qwen2.5-Coder-32B has 31.