Repository / RESOURCE PROFILE
Reproducible benchmarks for LLM output drift and agent replay in financial operations: DFAH-Bench, the installable dfah package, and a hands-on workshop
Reproducible benchmarks for LLM output drift and agent replay in financial operations: DFAH-Bench, the installable dfah package, and a hands-on workshop
Source: GitHub API · observed Oct 11, 2026. Stars and forks are recorded counts. Repository code licenses may differ from dataset or model-weight licenses.
References explain how this resource entered the collection. They do not represent an endorsement.
Agent Trading Arena probes numerical understanding of LLM-based trading agents in a zero-sum market simulation
InvestorBench evaluates LLM-based agents on financial decision-making tasks across multiple asset classes and markets
FinMTEB is a finance-specific massive text embedding benchmark with 64 datasets across seven financial NLP task types
