pixiu@parity

PIXIU: A Large Language Model, Instruction Data and Evaluation Benchmark for Finance. Total tasks: 435 across 29 financial NLP datasets.

← Back to Registry

Run this task

CLI:

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

Python:

from inspect_ai import eval
from inspect_harbor import pixiu_parity

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

Dataset information

Harbor registry	pixiu@parity
Inspect task	pixiu_parity
Version	parity
Samples	435
Paper	arxiv

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