Inception: Mercury 2.5

Text input Text output
Author's Description

Mercury 2.5 is the fastest reasoning LLM, and the latest diffusion LLM (dLLM) from Inception. Instead of generating tokens sequentially, Mercury 2.5 produces and refines multiple tokens in parallel, achieving...

Key Specifications
Cost
$$$$$
Context
260K
Released
Sep 08, 2026
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Stop Temperature Tool Choice Response Format Include Reasoning Reasoning Tools Max Tokens Structured Outputs
Features

This model supports the following features:

Reasoning Tools Response Format Structured Outputs
Performance Summary

Inception: Mercury 2.5, created on September 8, 2026, is positioned as a leading reasoning and diffusion LLM, notable for its parallel token generation and refinement. The model consistently ranks among the fastest available, achieving an Infinityth percentile in speed across eight benchmarks. It also offers highly competitive pricing, similarly ranking in the Infinityth percentile. Demonstrating exceptional operational stability, Mercury 2.5 boasts a 100% success rate across all benchmarks, indicating outstanding reliability with minimal technical failures. Performance across benchmarks reveals significant strengths in specific areas. Mercury 2.5 achieved perfect accuracy in Hallucinations (Baseline) and General Knowledge (Baseline), notably being the most accurate model at its respective price points and speeds for these categories. It also excelled in Coding (98.0% accuracy, 99th percentile) and Mathematics (98.0% accuracy, 99th percentile), showcasing robust capabilities in technical and quantitative domains. A notable weakness is its 0.0% accuracy in Instruction Following (Baseline), suggesting a critical area for improvement. While performing well in Reasoning (96.0% accuracy) and Ethics (99.0% accuracy), its Email Classification accuracy (98.0%) is more moderate. Overall, Mercury 2.5 stands out for its speed, cost-efficiency, reliability, and strong performance in knowledge-based and technical tasks, despite a significant gap in instruction following.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.04
Completion $0.15
Input Cache Read $0.004

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Inception
Inception | inception/mercury-2.5-20260908 260K $0.04 / 1M tokens $0.15 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by inception