Mercury 2
Inception · 1 configuration
Intelligence13.8
Input / 1M$0.25
Output / 1M$0.75
Context128K
Tariff: Inception · Source: artificial-analysis · USD per million tokens
Thinking configuration
What does more thinking buy you?
Compare this model’s measured thinking configurations.
Scroll the chart ↔ · tap a point for details
Efficient frontier: best measured score at each cost1 measured configurations · 0 without both values
Selected: thinking thinking · Click a point to switch configuration
Mercury 2 is an extremely fast reasoning LLM, and the first reasoning diffusion LLM (dLLM). Instead of generating tokens sequentially, Mercury 2 produces and refines multiple tokens in parallel, achieving...
Intelligence IndexArtificial Analysis composite index, not a percentage. Different versions of the index are not directly comparable.
13.8 pointsGPQA DiamondGraduate-level scientific questions. Measures specialist reasoning rather than web-search quality.
77.0 %Humanity’s Last ExamDifficult questions across many disciplines, compared under the same evaluation setup.
17.1 %SciCodeCode generation for scientific research problems.
37.7 %Terminal-Bench 2.1Agent tasks in a terminal environment. Results also depend on the evaluation harness.
27.3 %τ²-BenchTool use and interaction in assistance scenarios. Does not replace an evaluation of your integration.
70.8 %IFBenchVerifiable instruction following. A proxy for precision rather than a benchmark of creative style.
69.8 %Omniscience accuracyFactual accuracy in Artificial Analysis's Omniscience evaluation.
21.2 %Specifications from the catalog
Inputtext
Outputtext
Tool callingYes
Structured outputYes
Speed776.8 tok/s
First token4.32 s
Maximum output50K
Catalog listing date2026-03-04