Model comparison
DeepSeek-V2.5 (Sep 2024) vs Mercury 2
Mercury 2 is the stronger model overall, scoring 39.1 to 37.6 on the Noometry Index.
Last verified . 11 shared benchmarks.
DeepSeekSummary
- They share 11 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 1 category and Mercury 2 in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Mercury 2 leads 46.6 to 42.5.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Mercury 2 | |
|---|---|---|
| Provider | DeepSeek | Inception |
| Noometry Index | 37.6 | 39.1 |
| Released | 2024-09-06 | 2026-02-20 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 50K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $0.75 |
| Results tracked | 22 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mercury 2 leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Mercury 2: 33.5 (#255)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mercury 2 |
|---|---|---|
| LMArena Coding | 1309 | 1391 |
| Aider Polyglot | 17.8% | — |
| LMArena WebDev | — | 1171 |
| SciCode | — | 38.7% |
| WeirdML | — | 43.2% |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| ALE-Bench | — | 785.58 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Mercury 2: 23.8 (#170)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mercury 2 |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1362 |
| CritPt | — | 0.8% |
Math Not comparable
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Mercury 2: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mercury 2 |
|---|---|---|
| LMArena Math | 1288 | — |
Knowledge Mercury 2 leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Mercury 2: 36.2 (#172)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mercury 2 |
|---|---|---|
| LMArena Expert | 1266 | 1358 |
| Vectara Hallucination Rate | — | 12.3% |
Multilingual Mercury 2 leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Mercury 2: 46.6 (#157)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mercury 2 |
|---|---|---|
| LMArena Non-English | 1273 | 1331 |
| LMArena Chinese | 1318 | 1417 |
| LMArena Russian | 1289 | 1304 |
| LMArena French | 1289 | — |
| LMArena German | 1258 | — |
| LMArena Japanese | 1228 | — |
| LMArena Korean | 1209 | — |
| LMArena Spanish | 1248 | — |
Instruction Following Mercury 2 leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Mercury 2: 70.2 (#165)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mercury 2 |
|---|---|---|
| LMArena Instruction Following | 1280 | 1329 |
Long Context Too close to call
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Mercury 2: 40.5 (#154)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mercury 2 |
|---|---|---|
| LMArena Longer Query | 1301 | 1330 |
Writing & Preference Mercury 2 leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Mercury 2: 53.8 (#155)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mercury 2 |
|---|---|---|
| LMArena Text | 1294 | 1355 |
| LMArena Creative Writing | 1285 | 1289 |
| LMArena Multi-Turn | 1297 | 1358 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Mercury 2?
Mercury 2 is the stronger model overall, scoring 39.1 to 37.6 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Mercury 2 better for coding?
Mercury 2 scores higher on coding benchmarks: 33.5 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Mercury 2 share?
11 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Mercury 2 has 17.