blobfishai/ledgerbench-100

Finance

Professional

100 causal corporate-finance workflows over a closed ERP sandbox: provider-shaped MCP tools, distributed evidence, exact state transitions, and no LLM judge.

← Back to Registry

Run this task

CLI:

inspect eval inspect_harbor/blobfishai_ledgerbench_100 --model openai/gpt-5

Python:

from inspect_ai import eval
from inspect_harbor import blobfishai_ledgerbench_100

eval(blobfishai_ledgerbench_100(), model="openai/gpt-5")

Dataset information

Harbor registry	blobfishai/ledgerbench-100
Inspect task	`blobfishai_ledgerbench_100`
Latest digest	sha256:5e685691b3032c91489b8098947a08faabe9da131f6140c2ac54e1ac0092f693
Samples	100
Source	https://github.com/blobfishai/finance-world

See Task Parameters for the parameter set shared across all Harbor tasks.