Entity
RED-2400 – Solana DEX Rejected Trading Events Benchmark
RED-2400 is a public dataset of over 6,600 algorithmically-rejected Solana DEX trading events linked to post-rejection outcome labels, collected over 22 days in live production. It provides the first publicly available benchmark for evaluating DEX risk filter performance. The dataset has implications for DeFi quantitative trading, fraud detection, and emerging regulatory frameworks.
Importance: 60%Confidence: 75%Mentions: 1Updated: June 20, 2026
## RED-2400 – Solana DEX Rejected Trading Events Benchmark
### Overview
RED-2400 is a publicly released benchmark dataset of 6,660 algorithmically-rejected trading events from a live Solana decentralized-exchange (DEX) filter stack, observed over 22 calendar days from April 10 to May 2, 2026 (arXiv:2605.12151v2). The dataset links each rejection event to its post-rejection price-and-liquidity trajectory, enabling outcome-labelled analysis of DEX risk filters.
### Dataset Characteristics
- **169,123** forward-outcome observations
- **1,837** graveyard-tracker lifecycle snapshots
- **1,076** distinct mints in the rejection registry
- Outcome labels follow a defined classification schema tied to post-rejection price/liquidity behavior
- Continuously observed in a live production environment
### Strategic Relevance
- **Quantitative trading & DeFi risk management**: Provides the first publicly available labeled dataset for evaluating algorithmic rejection logic on Solana DEXs — directly useful for backtesting filter stacks and calibrating risk models.
- **Regulatory & compliance**: As DEX activity draws increasing regulatory scrutiny (SEC, CFTC), outcome-labelled rejection data may inform compliance frameworks distinguishing legitimate from manipulative order flow.
- **Market microstructure research**: The graveyard-tracker snapshots enable analysis of token lifecycle post-rejection, relevant to rug-pull detection and fraud identification.
- **Benchmark competition**: Establishes an empirical baseline against which future DEX filter models can be evaluated — analogous to benchmark datasets in traditional finance ML.
### Context
Solana DEX activity has grown substantially; filter stacks that reject potentially harmful trades (e.g., sandwich attacks, low-liquidity token manipulation) are critical infrastructure. Publicly benchmarking these systems is novel.