# Mercury 2 vs Qwen2.5-Coder-32B

> Mercury 2 is the stronger model overall, scoring 39.1 to 33.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/mercury-2-vs-qwen2-5-coder-32b
- Last updated: 2026-10-11
- Shared benchmarks: 11

## Summary

- They share 11 benchmarks with published results for both. Mercury 2 scores higher in 7 categories and Qwen2.5-Coder-32B in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mercury 2 leads 53.8 to 41.6.
- Mercury 2 is cheaper at $0.25 / $0.75 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- Mercury 2 accepts more context: 128K tokens versus 33K.
- Qwen2.5-Coder-32B has downloadable open weights; the other is API-only.

## Snapshot

| | Mercury 2 | Qwen2.5-Coder-32B |
|---|---|---|
| Provider | Inception | Alibaba (Qwen) |
| Noometry Index | 39.1 | 33.4 |
| Rank | 175 | 245 |
| Context | 128K | 33K |
| Input $/M | $0.25 | $0.66 |
| Output $/M | $0.75 | $1 |
| Weights | Proprietary | Open |

## Coding

- Mercury 2: 33.5 (#255)
- Qwen2.5-Coder-32B: 22.6 (#333)

| Benchmark | Mercury 2 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1391 | 1276 |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| LMArena WebDev | 1171 | — |
| SciCode | 38.7% | — |
| WeirdML | 43.2% | — |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| ALE-Bench | 785.58 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |

## Reasoning

- Mercury 2: 23.8 (#170)
- Qwen2.5-Coder-32B: 21.2 (#225)

| Benchmark | Mercury 2 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1362 | 1251 |
| CritPt | 0.8% | — |
| LiveBench Reasoning | — | 42.1% |
| LiveBench Data Analysis | — | 49.9% |
| Epoch Capabilities Index | — | 119.49 |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |

## Math

- Mercury 2: —
- Qwen2.5-Coder-32B: 33.3 (#204)

| Benchmark | Mercury 2 | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Math | — | 46.6% |
| LMArena Math | — | 1251 |
| GSM8K | — | 93% |

## Knowledge

- Mercury 2: 36.2 (#172)
- Qwen2.5-Coder-32B: 33.4 (#203)

| Benchmark | Mercury 2 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1358 | 1221 |
| Vectara Hallucination Rate | 12.3% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |

## Multilingual

- Mercury 2: 46.6 (#157)
- Qwen2.5-Coder-32B: 37.8 (#235)

| Benchmark | Mercury 2 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1331 | 1205 |
| LMArena Chinese | 1417 | 1222 |
| LMArena Russian | 1304 | 1228 |

## Instruction Following

- Mercury 2: 70.2 (#165)
- Qwen2.5-Coder-32B: 61.4 (#245)

| Benchmark | Mercury 2 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1329 | 1223 |
| LiveBench Instruction Following | — | 58.7% |

## Long Context

- Mercury 2: 40.5 (#154)
- Qwen2.5-Coder-32B: 38.0 (#208)

| Benchmark | Mercury 2 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1330 | 1251 |

## Writing & Preference

- Mercury 2: 53.8 (#155)
- Qwen2.5-Coder-32B: 41.6 (#240)

| Benchmark | Mercury 2 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1355 | 1230 |
| LMArena Creative Writing | 1289 | 1174 |
| LMArena Multi-Turn | 1358 | 1222 |
| LiveBench Language | — | 23.3% |

## FAQ

### Is Mercury 2 better than Qwen2.5-Coder-32B?

Mercury 2 is the stronger model overall, scoring 39.1 to 33.4 on the Noometry Index.

### Which is cheaper, Mercury 2 or Qwen2.5-Coder-32B?

Mercury 2 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.

### Is Mercury 2 or Qwen2.5-Coder-32B better for coding?

Mercury 2 scores higher on coding benchmarks: 33.5 versus 22.6 in the Noometry coding category.

### Which has the bigger context window?

Mercury 2 does, with 128K tokens against 33K.

### How many benchmarks do Mercury 2 and Qwen2.5-Coder-32B share?

11 benchmarks have published results for both models. Mercury 2 has 17 scored results on Noometry and Qwen2.5-Coder-32B has 31.