Batch Inference

Classify rows in a CSV

Run offline batch classification over a CSV file — label each row with a category, sentiment, or custom class using an open model, with results written back as a labeled CSV.

Estimated $0.001–$0.01 per row, depending on model size · Scales with row count — roughly 500–2,000 rows/min

What inputs you need

  • CSV file with a text column to classify
  • Label set / classification prompt
  • Model choice

What Badgr returns

  • Labeled CSV (original rows + predicted label + confidence)
  • Batch run log
  • Single receipt on completion

Recommended GPU routes

Estimated cost: $0.001–$0.01 per row, depending on model size · Estimated runtime: Scales with row count — roughly 500–2,000 rows/min

Example command

badgr run "python batch_classify.py --input rows.csv --labels labels.json" \
  --gpu A100_40GB \
  --max-cost 10 \
  --max-runtime 60

Common setup failures Badgr avoids

Job stops partway through a large CSV — input is chunked and progress checkpointed so a restart resumes instead of reprocessing.

Max-cost hit before the full file completes — estimate cost per row from a small sample run before submitting the full CSV.

Ready to run this?

Launch from the dashboard, CLI, or Compute API. Max-cost protection included.