#benchmarks

'A concise 2025 guide to AI agents covering what they are, where they work reliably, risks, architecture patterns, and evaluation strategies.'

VERINA introduces a holistic benchmark for evaluating LLMs on verifiable code generation, combining code, formal specifications, and proofs across diverse difficulty levels.

LifelongAgentBench introduces a novel benchmark for evaluating continuous learning in LLM-based agents, focusing on knowledge retention and adaptation across sequential tasks in dynamic environments.

Salesforce AI Research introduces innovative benchmarks, safety models, and architectures to build more reliable and capable AI agents for enterprise applications.