Google Cuts Gemma 4 31B API Pricing by 29% on Inputs and Outputs

Google has trimmed API pricing on Gemma 4 31B across the board, with a blended reduction of -29%. The numbers:

• Input: $0.14 → $0.10 per 1M tokens
• Output: $0.40 → $0.34 per 1M tokens

Does This Move the Needle for Your Workload?

For input-heavy use cases — RAG pipelines, long-context summarisation, document classification — the drop from $0.14 to $0.10 per million tokens is a meaningful 29% saving on that side of the bill. Output-heavy tasks (code generation, long-form drafting) see a smaller but still real cut: $0.40 down to $0.34.

As a rough sanity check: if you're pushing 100M input tokens a month, you've just saved $4. At 1B tokens, that's $40/month back in your budget — not transformative, but not nothing either.

Context

Gemma 4 31B sits in the mid-size open-weights tier, competing on price and accessibility rather than top-end benchmark scores. This cut keeps it competitive as smaller, cheaper models continue to improve and alternatives crowd the same price bracket. If you're already using Gemma 4 31B, no action required — you'll see lower costs automatically. If you've been comparing it against alternatives, the pricing gap has now shifted slightly in its favour.

Check current specs, benchmark comparisons, and the full price history on the Gemma 4 31B — live specs & price history page before committing to a workload migration.

Ezra, Scout AI Team