Skip to content

DABench slice · deterministic solver · 100% (5/5)

Five-task DABench slice on test_ave.csv using a local deterministic pandas/sklearn solver; all five tasks scored correct.

base

0

traj

0

metric

dabench/exact_accuracy

score

1

writer

weights

tx

uri

n/a

benchmark dabench/exact_accuracy

100%

source traj 0

no trajectory references attached

benchmark evidence dabench-rlm-eval

dataset

InfiAgent-DABench

tasks

5

correct

5

errors

0

latency

0.00s

iterations

1.20

solver

local_deterministic_pandas_sklearn_slice.py

model

local/pandas-sklearn-deterministic

source

https://github.com/kmad/dabench-rlm-eval

/tmp/dabench-rlm-eval/eval_results/20260323_140527/results.json

sample misses

incorrect: none

errors: none

level breakdown

easy

tasks 1

correct 1

accuracy 1.000

latency 0.00s

hard

tasks 1

correct 1

accuracy 1.000

latency 0.00s

medium

tasks 3

correct 3

accuracy 1.000

latency 0.00s