Noometry

Model comparison

DeepSeek-V3.1 vs Mercury 2

DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 39.1 on the Noometry Index.

Last verified . 13 shared benchmarks.

DeepSeek-V3.1 DeepSeek

42.8

Rank #108 Confirmed

Mercury 2 Inception

39.1

Rank #175 Confirmed

Summary

  • They share 13 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 6 categories and Mercury 2 in 1 category; 7 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where DeepSeek-V3.1 leads 43.7 to 36.2.
  • The biggest single-benchmark swing is Vectara Hallucination Rate: 5.5% for DeepSeek-V3.1 and 12.3% for Mercury 2.
  • Mercury 2 is cheaper at $0.25 / $0.75 per million input/output tokens, against $0.25 / $0.95 for DeepSeek-V3.1.
  • DeepSeek-V3.1 accepts more context: 164K tokens versus 128K.
  • DeepSeek-V3.1 has downloadable open weights; the other is API-only.

Side by side

DeepSeek-V3.1Mercury 2
ProviderDeepSeekInception
Noometry Index42.839.1
Released2025-08-212026-02-20
WeightsOpenProprietary
Context window164K128K
Max output8K50K
Input $ / M tokens$0.25$0.25
Output $ / M tokens$0.95$0.75
Results tracked2717

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding DeepSeek-V3.1 leads

DeepSeek-V3.1: 40.3 (#144), Mercury 2: 33.5 (#255)

BenchmarkDeepSeek-V3.1Mercury 2
WeirdML38.4%43.2%
LMArena Coding14171391
LMArena WebDev—1171
SciCode—38.7%
ALE-Bench—785.58

Reasoning DeepSeek-V3.1 leads

DeepSeek-V3.1: 27.9 (#110), Mercury 2: 23.8 (#170)

BenchmarkDeepSeek-V3.1Mercury 2
LMArena Hard Prompts14171362
SimpleBench40%—
Kagi LLM Benchmark53.2%—
CritPt—0.8%
DTBench82.7%—
LMCA24.3%—
Epoch Capabilities Index139.92—
ForecastBench58—

Math Not comparable

DeepSeek-V3.1: 38.9 (#122), Mercury 2: —

BenchmarkDeepSeek-V3.1Mercury 2
LMArena Math1420—

Knowledge DeepSeek-V3.1 leads

DeepSeek-V3.1: 43.7 (#90), Mercury 2: 36.2 (#172)

BenchmarkDeepSeek-V3.1Mercury 2
Vectara Hallucination Rate5.5%12.3%
LMArena Expert14051358

Multilingual DeepSeek-V3.1 leads

DeepSeek-V3.1: 51.6 (#106), Mercury 2: 46.6 (#157)

BenchmarkDeepSeek-V3.1Mercury 2
LMArena Non-English14001331
LMArena Chinese14691417
LMArena Russian14051304
LMArena French1447—
LMArena German1411—
LMArena Japanese1378—
LMArena Korean1337—
LMArena Spanish1431—

Instruction Following DeepSeek-V3.1 leads

DeepSeek-V3.1: 73.9 (#110), Mercury 2: 70.2 (#165)

BenchmarkDeepSeek-V3.1Mercury 2
LMArena Instruction Following14001329

Long Context Mercury 2 leads

DeepSeek-V3.1: 36.3 (#232), Mercury 2: 40.5 (#154)

BenchmarkDeepSeek-V3.1Mercury 2
LMArena Longer Query14221330
Fiction.LiveBench52.8%—

Writing & Preference DeepSeek-V3.1 leads

DeepSeek-V3.1: 60.3 (#98), Mercury 2: 53.8 (#155)

BenchmarkDeepSeek-V3.1Mercury 2
LMArena Text14201355
LMArena Creative Writing14011289
LMArena Multi-Turn14081358
EQ-Bench Creative Writing1436—

Frequently asked questions

Is DeepSeek-V3.1 better than Mercury 2?

DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 39.1 on the Noometry Index.

Which is cheaper, DeepSeek-V3.1 or Mercury 2?

Mercury 2 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; DeepSeek-V3.1 lists at $0.25 and $0.95.

Is DeepSeek-V3.1 or Mercury 2 better for coding?

DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 33.5 in the Noometry coding category.

Which has the bigger context window?

DeepSeek-V3.1 does, with 164K tokens against 128K.

How many benchmarks do DeepSeek-V3.1 and Mercury 2 share?

13 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Mercury 2 has 17.

Related comparisons

Go deeper