Is Step 3.7 Flash Thinking or Gemma 4 31B Garnet cheaper?

Gemma 4 31B Garnet is cheaper on output tokens ($0.31 vs $1.15 per 1M).

Which has the larger context window, Step 3.7 Flash Thinking or Gemma 4 31B Garnet?

Gemma 4 31B Garnet has the larger context window (262K tokens).

Step 3.7 Flash Thinking vs Gemma 4 31B Garnet

Gemma 4 31B Garnet is cheaper on output tokens, while Gemma 4 31B Garnet offers a larger context window. Choose Step 3.7 Flash Thinking or Gemma 4 31B Garnet based on the trade-off between cost, context, and the benchmarks that matter for your use case.

Spec	Step 3.7 Flash Thinking	Gemma 4 31B Garnet
Provider	NanoGPT	NanoGPT
Input / 1M tokens	$0.20	$0.31
Output / 1M tokens	$1.15	$0.31
Context window	256K	262K
Parameters	—	—
Open weights	No	No
Released	May 2026	May 2026

Step 3.7 Flash Thinking details →Gemma 4 31B Garnet details →

FAQ

Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.