Inception Proprietary Released Feb 24, 2026 Updated
Mercury 2
ToolsGuide Scores, rankings, test results and price.
Model summary
Summary
Mercury 2 is Inception's model, released Feb 24, 2026. It's ranked in 2 categories and does best at text: #190 of 416, with a ToolsGuide Score of 246 against a median of 235.
It also ranks #139 of 147 for coding (171).
It costs $0.25 per million input tokens and $0.75 per million output tokens, below the median text model ($0.43).
It has published results on 6 tests.
Specifications
- Creator
- Inception
- Licence
- Proprietary
- Released
- Feb 24, 2026
- Context window
- 128K tokens
- Input price
- $0.25 / 1M
- Output price
- $0.75 / 1M
- Tests reported
- 6
- Settings ranked
- mercury-2, Mercury 2
Model comparison
Against the ten best text models.
Text score
ToolsGuide Score out of 1,000 · Higher is better
Coding score
ToolsGuide Score out of 1,000 · Higher is better
Input price
USD per 1M input tokens · Lower is better
Leaderboards
Where Mercury 2 sits among the top 20 in each category.
Text
#190 of 416 · ToolsGuide Score out of 1,000, where 1,000 is a flawless model
Coding
#139 of 147 · ToolsGuide Score out of 1,000, where 1,000 is a flawless model
By task
Scored from votes on each kind of prompt alone, on the same 1,000-point scale.
Score against price
Text · the 10 priced models ranked closest · up and to the left is better value
Text
mercury-2 · hover a bar for its rank
Coding
Mercury 2 · hover a bar for its rank
Test results
Every test Mercury 2 has a published result on, out of 100%, with its rank among the models that took it.
GPQA
#131 / 256Graduate-level science questions written to be hard to look up.
74.0%
Self-reported
AIME 2025
#44 / 123Competition maths problems from the 2025 American Invitational Mathematics Exam.
91.1%
Self-reported
LiveCodeBench
#27 / 76LiveCodeBench is a holistic and contamination-free evaluation benchmark for large language models for code. It continuously collects new problems from programming contests (LeetCode, AtCoder, CodeForces) and evaluates four different scenarios: code generation, self-repair, code execution, and test output prediction. Problems are annotated with release dates to enable evaluation on unseen problems released after a model's training cutoff.
67.0%
Self-reported
IFBench
#26 / 44Instruction Following Benchmark evaluating model's ability to follow complex instructions
71.0%
Self-reported
SciCode
#22 / 30SciCode is a research coding benchmark curated by scientists that challenges language models to code solutions for scientific problems. It contains 338 subproblems decomposed from 80 challenging main problems across 16 natural science sub-fields including mathematics, physics, chemistry, biology, and materials science. Problems require knowledge recall, reasoning, and code synthesis skills.
38.0%
Self-reported
Tau2 Airline
#21 / 25TAU2 airline domain benchmark for evaluating conversational agents in dual-control environments where both AI agents and users interact with tools in airline customer service scenarios. Tests agent coordination, communication, and ability to guide user actions in tasks like flight booking, modifications, cancellations, and refunds.
53.0%
Self-reported