Model comparison

gpt-oss-20b vs Kimi K2.7 Code

Pricing, context window and real answers (September 2026)

gpt-oss-20b (OpenAI) and Kimi K2.7 Code (Moonshot), compared as you can actually call them on FastMetal. Both are served from the same OpenAI-compatible endpoint and API key; switching is a change to the model string.

Specs and pricing

openai logogpt-oss-20bmoonshotai logoKimi K2.7 Code
ProviderOpenAIMoonshot
Input (per 1M tokens)¥5.36¥55
Output (per 1M tokens)¥25.02¥530
Typical cost (1,000 in + 500 out tokens × 1,000 calls)¥18¥320
Context window131,072 tokens262,144 tokens
Release date8/5/20256/12/2026
Input modalitiestexttext, image
Throughput (p50)
Arena · overall#252 · ELO 1,317Unranked
Arena · Japanese#179 · ELO 1,256Unranked
Arena · coding#232 · ELO 1,369Unranked

Prices are FastMetal's yen rates before tax. Arena ranks are as of the latest leaderboard scrape.

Which should you pick?

  • On price, gpt-oss-20b: ¥25.02 per 1M output tokens, 95% less than Kimi K2.7 Code.
  • For long documents, Kimi K2.7 Code: a 262,144-token context window against 131,072 for gpt-oss-20b.
  • If in doubt, try both on the same key. Switching is a change to the model string, and each is billed at its own rate.

Frequently asked questions

Which is cheaper, gpt-oss-20b or Kimi K2.7 Code?
Per 1M output tokens, gpt-oss-20b is ¥25.02 and Kimi K2.7 Code is ¥530 on FastMetal (yen, before tax), so gpt-oss-20b is cheaper.
How do the context windows of gpt-oss-20b and Kimi K2.7 Code compare?
gpt-oss-20b takes 131,072 tokens; Kimi K2.7 Code takes 262,144.
Can I use gpt-oss-20b and Kimi K2.7 Code with the same API key?
Yes. On FastMetal's OpenAI-compatible endpoint, switch by passing "gpt-oss-20b" or "japan-kimi-k2.7-code" as the model. Each is billed at its own rate from the same prepaid balance.

Try both on one API key

Create an account and add credit to call gpt-oss-20b and Kimi K2.7 Code from the browser chat and the API. No monthly fee.

More comparisons with gpt-oss-20b

More comparisons with Kimi K2.7 Code