Is Qwen/Qwen2.5-72B-Instruct or moonshotai/Kimi-K2.5 cheaper?

Qwen/Qwen2.5-72B-Instruct is cheaper on output tokens ($0.59 vs $2.25 per 1M).

Which has the larger context window, Qwen/Qwen2.5-72B-Instruct or moonshotai/Kimi-K2.5?

moonshotai/Kimi-K2.5 has the larger context window (262K tokens).

Qwen/Qwen2.5-72B-Instruct vs moonshotai/Kimi-K2.5

Qwen/Qwen2.5-72B-Instruct is cheaper on output tokens, while moonshotai/Kimi-K2.5 offers a larger context window. Choose Qwen/Qwen2.5-72B-Instruct or moonshotai/Kimi-K2.5 based on the trade-off between cost, context, and the benchmarks that matter for your use case.

Spec	Qwen/Qwen2.5-72B-Instruct	moonshotai/Kimi-K2.5
Provider	SiliconFlow	SiliconFlow
Input / 1M tokens	$0.59	$0.45
Output / 1M tokens	$0.59	$2.25
Context window	33K	262K
Parameters	—	1T
Open weights	No	Yes
Released	Sep 2024	Jan 2026

Qwen/Qwen2.5-72B-Instruct details →moonshotai/Kimi-K2.5 details →

FAQ

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