◫ Curated ranking

Cheapest AI models with batch API

Batch API endpoints offer ~50% off in exchange for a 24-hour SLA. For non-interactive workloads — evals, document processing, content generation — this is the single biggest cost lever available. Ranked by effective input price (input × 0.5) ascending.

Methodology

How this ranking is built

Active models accessible via the batch API with a published input price. Ranked by what you'd actually pay through batch — the typical batch discount factored in — from lowest cost upward.

#	Model	Provider	Effective batch input	Lifecycle	Verified
1	Gemini 2.5 Flash-Lite Google Gemini · Active	GG Google Gemini	$0.05/M (batch)	Active	· 1h ago
2	Gemini 2.5 Flash Google Gemini · Active	GG Google Gemini	$0.15/M (batch)	Active	· 1h ago
3	Amazon Nova Pro AWS Bedrock · Active	AB AWS Bedrock	$0.4/M (batch)	Active	· 1h ago
4	Gemini 2.5 Pro Google Gemini · Active	GG Google Gemini	$0.63/M (batch)	Active	· 1h ago
5	Gemini 3.5 Flash Google Gemini · Active	GG Google Gemini	$0.75/M (batch)	Active	· 1h ago

Refreshed as we re-verify each model against its provider's official docs. Every value links to the page it came from. Last updated Jun 15, 2026 18:02 UTC.