M

Mercury 2.5

Mercuryv2.5Current
Released September 8, 2026
Context260K tokens
Price (In / Out)$0.04/M / $0.15/M
CategoryLarge Language Model

About Mercury 2.5

Mercury 2.5 is a proprietary reasoning model from Inception listed by Artificial Analysis. The model page reports a 260,000-token context window, high output speed, and low listed token pricing for a reasoning-capable model.

Capabilities

textreasoninglong context

Input Modalities

text

Output Modalities

text

Technical Details

API Identifier
mercury-2-5
Category
Large Language Model
Context Window
260,000 tokens

Tags

mercuryinceptionreasoning-modellong-contextartificial-analysisproprietary

Benchmarks

Performance scores for Mercury 2.5 across standard benchmarks.

Artificial Analysis Intelligence Indexhttps://artificialanalysis.ai/models/mercury-2-5 · Sep 2026
12
AA median output speed tokens/sechttps://artificialanalysis.ai/models/mercury-2-5 · Sep 2026
770.4
AA median time to first chunk secondshttps://artificialanalysis.ai/models/mercury-2-5 · Sep 2026
3.1

Pricing

Token pricing for Mercury 2.5 API usage.

Input Tokens

$0.04/M

per million tokens

Output Tokens

$0.15/M

per million tokens

Inception's direct model page currently advertises launch pricing of $0.04/M input and $0.15/M output, with $0.004/M cached input. Standard prices shown alongside are $0.20/M input, $0.75/M output and $0.02/M cached input. The discount duration is not specified; reverify before budgeting.

Competing Models

Same pricing tier — direct alternatives to Mercury 2.5

Efficiency