Model comparison
Qwen2.5 7B Instruct vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 32.9 to 28.6 on the Noometry Index. Qwen2.5 7B Instruct costs 2.4× less per token, which makes it the better buy when Qwen2.5-Coder-32B's lead doesn't matter for your workload.
Last verified . 4 shared benchmarks.
Alibaba (Qwen)Summary
- They share 4 benchmarks with published results for both. Qwen2.5 7B Instruct scores higher in 3 categories and Qwen2.5-Coder-32B in 3 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2.5-Coder-32B leads 30.8 to 11.8.
- The biggest single-benchmark swing is BigCodeBench Complete: 46.1% for Qwen2.5 7B Instruct and 58% for Qwen2.5-Coder-32B.
- Qwen2.5 7B Instruct is cheaper at $0.17 / $0.70 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- Qwen2.5 7B Instruct accepts more context: 131K tokens versus 33K.
Side by side
| Qwen2.5 7B Instruct | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 28.6 | 32.9 |
| Released | 2024-09 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 131K | 33K |
| Max output | 8K | 29K |
| Input $ / M tokens | $0.17 | $0.66 |
| Output $ / M tokens | $0.70 | $1 |
| Results tracked | 15 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 7B Instruct leads
Qwen2.5 7B Instruct: 35.9 (#204), Qwen2.5-Coder-32B: 22.4 (#328)
| Benchmark | Qwen2.5 7B Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| BigCodeBench Instruct | 37.6% | 49% |
| BigCodeBench Complete | 46.1% | 58% |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| LiveBench Coding | — | 56.9% |
| LMArena Coding | — | 1276 |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Agentic & Tool Use Not comparable
Qwen2.5 7B Instruct: 23.5 (#117), Qwen2.5-Coder-32B: —
| Benchmark | Qwen2.5 7B Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| BALROG | 7.8% | — |
Reasoning Qwen2.5-Coder-32B leads
Qwen2.5 7B Instruct: 14.3 (#317), Qwen2.5-Coder-32B: 20.4 (#220)
| Benchmark | Qwen2.5 7B Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| Epoch Capabilities Index | 118.44 | 119.41 |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 42.1% |
| LMArena Hard Prompts | — | 1251 |
| DTBench | 47.7% | — |
| LiveBench Data Analysis | — | 49.9% |
| LMCA | 6.4% | — |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math Qwen2.5-Coder-32B leads
Qwen2.5 7B Instruct: 11.8 (#300), Qwen2.5-Coder-32B: 30.8 (#203)
| Benchmark | Qwen2.5 7B Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.5% | — |
| Omni-MATH | 29.4% | — |
| LiveBench Math | — | 46.6% |
| LMArena Math | — | 1251 |
| GSM8K | — | 93% |
Knowledge Qwen2.5-Coder-32B leads
Qwen2.5 7B Instruct: 17.0 (#281), Qwen2.5-Coder-32B: 32.9 (#200)
| Benchmark | Qwen2.5 7B Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| MMLU | 72.9% | 79.1% |
| GPQA Diamond | 35.5% | — |
| MMLU-Pro | 53.9% | — |
| GPQA (HELM) | 34.1% | — |
| LMArena Expert | — | 1221 |
| ARC (AI2) Challenge | — | 70.5% |
Multilingual Not comparable
Qwen2.5 7B Instruct: —, Qwen2.5-Coder-32B: 38.0 (#230)
| Benchmark | Qwen2.5 7B Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | — | 1205 |
| LMArena Chinese | — | 1222 |
| LMArena Russian | — | 1228 |
Instruction Following Qwen2.5 7B Instruct leads
Qwen2.5 7B Instruct: 63.4 (#226), Qwen2.5-Coder-32B: 61.6 (#240)
| Benchmark | Qwen2.5 7B Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Instruction Following | — | 58.7% |
| IFEval | 74.1% | — |
| LMArena Instruction Following | — | 1223 |
Long Context Not comparable
Qwen2.5 7B Instruct: —, Qwen2.5-Coder-32B: 37.9 (#204)
| Benchmark | Qwen2.5 7B Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | — | 1251 |
Writing & Preference Qwen2.5 7B Instruct leads
Qwen2.5 7B Instruct: 48.8 (#194), Qwen2.5-Coder-32B: 41.8 (#235)
| Benchmark | Qwen2.5 7B Instruct | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | — | 1230 |
| LMArena Creative Writing | — | 1174 |
| WildBench | 73.1% | — |
| LMArena Multi-Turn | — | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Qwen2.5 7B Instruct better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 32.9 to 28.6 on the Noometry Index. Qwen2.5 7B Instruct costs 2.4× 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, Qwen2.5 7B Instruct or Qwen2.5-Coder-32B?
Qwen2.5 7B Instruct is cheaper. It lists at $0.17 per million input tokens and $0.70 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is Qwen2.5 7B Instruct or Qwen2.5-Coder-32B better for coding?
Qwen2.5 7B Instruct scores higher on coding benchmarks: 35.9 versus 22.4 in the Noometry coding category.
Which has the bigger context window?
Qwen2.5 7B Instruct does, with 131K tokens against 33K.
How many benchmarks do Qwen2.5 7B Instruct and Qwen2.5-Coder-32B share?
4 benchmarks have published results for both models. Qwen2.5 7B Instruct has 15 scored results on Noometry and Qwen2.5-Coder-32B has 31.