Model comparison
Deepseek Coder v2 vs Mercury 2
Mercury 2 is the stronger model overall, scoring 38.8 to 35.3 on the Noometry Index.
Last verified . 11 shared benchmarks.
DeepSeekSummary
- They share 11 benchmarks with published results for both. Deepseek Coder v2 scores higher in 1 category and Mercury 2 in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mercury 2 leads 54.0 to 38.4.
- Deepseek Coder v2 has downloadable open weights; the other is API-only.
Side by side
| Deepseek Coder v2 | Mercury 2 | |
|---|---|---|
| Provider | DeepSeek | Inception |
| Noometry Index | 35.3 | 38.8 |
| Released | 2024-06-17 | 2026-02-20 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 50K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $0.75 |
| Results tracked | 24 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Deepseek Coder v2 leads
Deepseek Coder v2: 37.5 (#178), Mercury 2: 33.2 (#250)
| Benchmark | Deepseek Coder v2 | Mercury 2 |
|---|---|---|
| LMArena Coding | 1251 | 1391 |
| LMArena WebDev | — | 1170 |
| SciCode | — | 38.7% |
| WeirdML | — | 43.2% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| ALE-Bench | — | 785.58 |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Reasoning Too close to call
Deepseek Coder v2: 22.6 (#175), Mercury 2: 23.0 (#170)
| Benchmark | Deepseek Coder v2 | Mercury 2 |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1362 |
| CritPt | — | 0.8% |
| WinoGrande | 83.7% | — |
Math Not comparable
Deepseek Coder v2: 32.3 (#188), Mercury 2: —
| Benchmark | Deepseek Coder v2 | Mercury 2 |
|---|---|---|
| LMArena Math | 1241 | — |
| GSM8K | 94.5% | — |
Knowledge Mercury 2 leads
Deepseek Coder v2: 31.7 (#209), Mercury 2: 35.6 (#170)
| Benchmark | Deepseek Coder v2 | Mercury 2 |
|---|---|---|
| LMArena Expert | 1181 | 1358 |
| Vectara Hallucination Rate | — | 12.3% |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual Mercury 2 leads
Deepseek Coder v2: 36.4 (#235), Mercury 2: 46.8 (#153)
| Benchmark | Deepseek Coder v2 | Mercury 2 |
|---|---|---|
| LMArena Non-English | 1182 | 1331 |
| LMArena Chinese | 1201 | 1416 |
| LMArena Russian | 1188 | 1303 |
| LMArena French | 1184 | — |
| LMArena German | 1164 | — |
| LMArena Japanese | 1126 | — |
| LMArena Korean | 1104 | — |
| LMArena Spanish | 1153 | — |
Instruction Following Mercury 2 leads
Deepseek Coder v2: 61.9 (#237), Mercury 2: 70.4 (#161)
| Benchmark | Deepseek Coder v2 | Mercury 2 |
|---|---|---|
| LMArena Instruction Following | 1180 | 1328 |
Long Context Mercury 2 leads
Deepseek Coder v2: 37.0 (#219), Mercury 2: 40.4 (#149)
| Benchmark | Deepseek Coder v2 | Mercury 2 |
|---|---|---|
| LMArena Longer Query | 1219 | 1329 |
Writing & Preference Mercury 2 leads
Deepseek Coder v2: 38.4 (#249), Mercury 2: 54.0 (#151)
| Benchmark | Deepseek Coder v2 | Mercury 2 |
|---|---|---|
| LMArena Text | 1191 | 1355 |
| LMArena Creative Writing | 1120 | 1289 |
| LMArena Multi-Turn | 1176 | 1358 |
Frequently asked questions
Is Deepseek Coder v2 better than Mercury 2?
Mercury 2 is the stronger model overall, scoring 38.8 to 35.3 on the Noometry Index.
Is Deepseek Coder v2 or Mercury 2 better for coding?
Deepseek Coder v2 scores higher on coding benchmarks: 37.5 versus 33.2 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and Mercury 2 share?
11 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Mercury 2 has 17.