Is Llama 3.1 70B Celeste v0.1 or Gemma 4 31B Garnet cheaper?

Gemma 4 31B Garnet is cheaper on output tokens ($0.31 vs $0.49 per 1M).

Which has the larger context window, Llama 3.1 70B Celeste v0.1 or Gemma 4 31B Garnet?

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

Llama 3.1 70B Celeste v0.1 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 Llama 3.1 70B Celeste v0.1 or Gemma 4 31B Garnet based on the trade-off between cost, context, and the benchmarks that matter for your use case.

Spec	Llama 3.1 70B Celeste v0.1	Gemma 4 31B Garnet
Provider	NanoGPT	NanoGPT
Input / 1M tokens	$0.49	$0.31
Output / 1M tokens	$0.49	$0.31
Context window	16K	262K
Parameters	—	—
Open weights	No	No
Released	Jul 2024	May 2026

Llama 3.1 70B Celeste v0.1 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.