Is Kimi K2 Instruct or GPT-OSS 20B cheaper?

GPT-OSS 20B is cheaper on output tokens ($0.14 vs $1.90 per 1M).

Which has the larger context window, Kimi K2 Instruct or GPT-OSS 20B?

GPT-OSS 20B has the larger context window (64K tokens).

Kimi K2 Instruct vs GPT-OSS 20B

GPT-OSS 20B is cheaper on output tokens, while GPT-OSS 20B offers a larger context window. Choose Kimi K2 Instruct or GPT-OSS 20B based on the trade-off between cost, context, and the benchmarks that matter for your use case.

Spec	Kimi K2 Instruct	GPT-OSS 20B
Provider	IO.NET	IO.NET
Input / 1M tokens	$0.39	$0.03
Output / 1M tokens	$1.90	$0.14
Context window	33K	64K
Parameters	—	21B
Open weights	No	Yes
Released	Sep 2024	Dec 2024

Kimi K2 Instruct details →GPT-OSS 20B details →

FAQ

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