Incomplete14%Harbor nativeAgentic / Tool Use

erp-bench

ERP-Bench is the Odoo 19 benchmark used in the Anchor paper, "Preventing Artifact Drift in Agent Benchmark Generation." It contains 300 long-horizon procurement and manufacturing tasks generated from a single CP-SAT-backed specification.

Homepage

How to run

ltb · erp-bench

$ npm i -g lettokenburn-cli
$ ltb bench run agentic-labs/erp-bench@1 --benchmark erp-bench -a claude-code -m anthropic/claude-opus-4-8
$ ltb bench push

Harbor-native: you can also call Harbor directly.

harbor

$ harbor run -d agentic-labs/erp-bench@1 -a claude-code -m anthropic/claude-opus-4-8

Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs

Scores

0 sourced entries

No sourced scores recorded yet.

Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.

Reproduced by the community

All runs

About

agentic-labs/erp-bench

Facts

Metric: N/A
Scale: 0–100
Scoring: N/A
Format: Agentic / environment
Samples: N/A
Difficulty: Medium
Saturation: Unknown
Languages: en
Modalities: text
Industries: General
Released: N/A
HF dataset: N/A

Data completeness

Incomplete14%

Incomplete: Fewer than 40% of the key fields are known. Treat with care.

Missing:

Paper link
Dataset or repository link
License
Sample count
Release date
Primary metric

Confirmed by 1 source channel. Fields are extracted automatically and can be wrong.

Report a correction ([email protected])

License

License N/Alink-only

Citation. Cite the original authors via the links above.

Index record

Review status: auto published

Extraction confidence: 85%

Updated: Oct 1, 2026

JSON record

How to run

Scores

Reproduced by the community

About

agentic-labs/erp-bench

Related benchmarks

KiranaBench