Model comparison

DeepSeek-V3.2-Exp vs Mercury 2.5

DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.3 to 33.5 on the Noometry Index. Mercury 2.5 costs 4.3× less per token, which makes it the better buy when DeepSeek-V3.2-Exp's lead doesn't matter for your workload.

Last verified October 10, 2026. 3 shared benchmarks.

DeepSeek-V3.2-Exp

DeepSeek

44.3

Rank #78 Confirmed

Mercury 2.5 Inception

33.5

Rank #242 Reported

Summary

They share 3 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 2 categories and Mercury 2.5 in 1 category; 2 gaps are clear of the uncertainty.
The widest gap is in math, where DeepSeek-V3.2-Exp leads 41.7 to 23.3.
The biggest single-benchmark swing is ProofBench: 8% for DeepSeek-V3.2-Exp and 3% for Mercury 2.5.
Mercury 2.5 is cheaper at $0.04 / $0.15 per million input/output tokens, against $0.26 / $0.38 for DeepSeek-V3.2-Exp.
Mercury 2.5 accepts more context: 260K tokens versus 164K.
DeepSeek-V3.2-Exp has downloadable open weights; the other is API-only.

Side by side

DeepSeek-V3.2-Exp and Mercury 2.5 specifications
	DeepSeek-V3.2-Exp	Mercury 2.5
Provider	DeepSeek	Inception
Noometry Index	44.3	33.5
Released	2025-09-29	2026-09-08
Weights	Open	Proprietary
Context window	164K	260K
Max output	66K	66K
Input $ / M tokens	$0.26	$0.04
Output $ / M tokens	$0.38	$0.15
Results tracked	49	4

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

Category by category

Coding DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 46.5 (#65), Mercury 2.5: 39.5 (#156)

Coding benchmarks
Benchmark	DeepSeek-V3.2-Exp	Mercury 2.5
SciCode	38.9%	38.5%
SWE-bench Verified (bash only)	70%	—
Aider Polyglot	74.2%	—
LMArena WebDev	1362	—
SWE-bench Multilingual	59%	—
WeirdML	39.5%	—
LMArena Coding	1454	—
ALE-Bench	—	301.65

Agentic & Tool Use Not comparable

DeepSeek-V3.2-Exp: 32.7 (#59), Mercury 2.5: —

Agentic & Tool Use benchmarks
Benchmark	DeepSeek-V3.2-Exp	Mercury 2.5
Terminal-Bench	39.6%	—
APEX-Agents	21.3%	—
Berkeley Function Calling Leaderboard	56.7%	—
TheAgentCompany	42.9%	—
Vending-Bench 2	1,034	—

Reasoning Too close to call

DeepSeek-V3.2-Exp: 22.1 (#208), Mercury 2.5: 22.4 (#193)

Reasoning benchmarks
Benchmark	DeepSeek-V3.2-Exp	Mercury 2.5
CritPt	2.9%	0%
ARC-AGI-2	4%	—
Kagi LLM Benchmark	52.2%	—
NYT Connections (extended)	36.7%	—
ARC-AGI-1	57%	—
Chess Puzzles	14%	—
Thematic Generalization	65%	—
LMArena Hard Prompts	1434	—
DTBench	87.7%	—
LMCA	29.1%	—
Epoch Capabilities Index	146.27	—

Math DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 41.7 (#87), Mercury 2.5: 23.3 (#272)

Math benchmarks
Benchmark	DeepSeek-V3.2-Exp	Mercury 2.5
ProofBench	8%	3%
MathArena Final-Answer Competitions	57.7%	—
OTIS Mock AIME 2024-2025	87.8%	—
LMArena Math	1435	—
FrontierMath (Feb 2025 set)	22.1%	—
FrontierMath Tier 4 (v1)	2.1%	—

Knowledge Not comparable

DeepSeek-V3.2-Exp: 51.7 (#66), Mercury 2.5: —

Knowledge benchmarks
Benchmark	DeepSeek-V3.2-Exp	Mercury 2.5
GPQA Diamond	83.4%	—
Vectara Hallucination Rate	5.3%	—
LMArena Expert	1436	—

Multilingual Not comparable

DeepSeek-V3.2-Exp: 52.2 (#90), Mercury 2.5: —

Multilingual benchmarks
Benchmark	DeepSeek-V3.2-Exp	Mercury 2.5
LMArena Non-English	1409	—
LMArena Chinese	1461	—
LMArena French	1433	—
LMArena German	1440	—
LMArena Japanese	1374	—
LMArena Korean	1371	—
LMArena Russian	1424	—
LMArena Spanish	1440	—

Instruction Following Not comparable

DeepSeek-V3.2-Exp: 74.5 (#93), Mercury 2.5: —

Instruction Following benchmarks
Benchmark	DeepSeek-V3.2-Exp	Mercury 2.5
LMArena Instruction Following	1413	—

Long Context Not comparable

DeepSeek-V3.2-Exp: 47.6 (#16), Mercury 2.5: —

Long Context benchmarks
Benchmark	DeepSeek-V3.2-Exp	Mercury 2.5
Fiction.LiveBench	83.3%	—
CL-bench	13.2%	—
CL-bench Life	9.5%	—
LMArena Longer Query	1428	—

Writing & Preference Not comparable

DeepSeek-V3.2-Exp: 62.4 (#77), Mercury 2.5: —

Writing & Preference benchmarks
Benchmark	DeepSeek-V3.2-Exp	Mercury 2.5
LMArena Text	1425	—
LMArena Creative Writing	1403	—
EQ-Bench Creative Writing	1515	—
LMArena Multi-Turn	1427	—

Frequently asked questions

Is DeepSeek-V3.2-Exp better than Mercury 2.5?

Which is cheaper, DeepSeek-V3.2-Exp or Mercury 2.5?

Mercury 2.5 is cheaper. It lists at $0.04 per million input tokens and $0.15 per million output tokens; DeepSeek-V3.2-Exp lists at $0.26 and $0.38.

Is DeepSeek-V3.2-Exp or Mercury 2.5 better for coding?

DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.5 versus 39.5 in the Noometry coding category.

Which has the bigger context window?

Mercury 2.5 does, with 260K tokens against 164K.

How many benchmarks do DeepSeek-V3.2-Exp and Mercury 2.5 share?

3 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 49 scored results on Noometry and Mercury 2.5 has 4.