# Gemma 4 31B API prices

Gemma 4 31B (gemma) — 262k context, 262k max output.
5 metered per-token offers from 6 provider(s), in USD per 1M tokens.

| Provider | Input $/1M | Output $/1M | Cache read $/1M | Context |
| --- | --- | --- | --- | --- |
| Weights & Biases | $0.1 | $0.34 | $0.1 | 262k |
| NanoGPT | $0.1 | $0.45 | $0.05 | 262k |
| NovitaAI | $0.14 | $0.4 | — | 262k |
| Neuralwatt | $0.144 | $0.42 | $0.0144 | 262k |
| NanoGPT | $0.4 | $1 | $0.4 | 262k |

## Summary

- Cheapest input: $0.1 per 1M tokens (Weights & Biases)
- Cheapest output: $0.34 per 1M tokens (Weights & Biases)
- First-party: not listed separately
- Free offers: OpenRouter, Poe
- Subscription plans (not per-token): none
- Released: 2026-04-02
- Inputs: image, text, video

HTML page: https://llmprice.gitlab.io/models/gemma-4-31b/
Full dataset: https://llmprice.gitlab.io/data/catalog.json