Home / Tools / Batch pricing

LLM batch pricing calculator

Compare standard vs Batch API cost for the same token profile. Batch usually trades higher latency for lower rates. Use it for offline jobs, not chat UIs. Same pricing engine as the main calculator.

Full cost calculator · Cache savings · Prompt cost guide

Model (batch pricing published)

OpenAI · standard $0.500/1M / $1.50/1M · batch $0.250/1M / $0.750/1M

Input tokens / job

Output tokens / job

Batch jobs / month

Standard request: $0.004768
Batch request: $0.002384
Saved / job: $0.002384 (50.0%)
Saved / month: $23.84

Uses published batch input/output rates when present; otherwise falls back to ~50% of standard (same rule as the main calculator). Batch does not change token counts, only the rate. Verified 2026-08-02.

When to use batch

Good for offline evaluation, report generation, embeddings-style bulk text jobs, and anything that can wait. Skip it for interactive chat where latency matters.

Open GPT-3.5 Turbo with Batch on the calculator · Cache savings · LLM pricing guide