Model comparison

Kimi K2.7 Code vs LLM-jp-3.1 8x13B instruct4

Pricing, context window and real answers (August 2026)

Kimi K2.7 Code (Moonshot) and LLM-jp-3.1 8x13B instruct4 (Llm Jp), 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

moonshotai logoKimi K2.7 CodeLLM-jp-3.1 8x13B instruct4
ProviderMoonshotLlm Jp
Input (per 1M tokens)¥55¥16
Output (per 1M tokens)¥530¥79
Typical cost (1,000 in + 500 out tokens × 1,000 calls)¥320¥56
Context window262,144 tokens4,096 tokens
Release date6/12/20263/21/2026
Input modalitiestext, imagetext
Arena · overallUnrankedUnranked
Arena · JapaneseUnrankedUnranked
Arena · codingUnrankedUnranked

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

Which should you pick?

  • On price, LLM-jp-3.1 8x13B instruct4: ¥79 per 1M output tokens, 85% less than Kimi K2.7 Code.
  • For long documents, Kimi K2.7 Code: a 262,144-token context window against 4,096 for LLM-jp-3.1 8x13B instruct4.
  • 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, Kimi K2.7 Code or LLM-jp-3.1 8x13B instruct4?
Per 1M output tokens, Kimi K2.7 Code is ¥530 and LLM-jp-3.1 8x13B instruct4 is ¥79 on FastMetal (yen, before tax), so LLM-jp-3.1 8x13B instruct4 is cheaper.
How do the context windows of Kimi K2.7 Code and LLM-jp-3.1 8x13B instruct4 compare?
Kimi K2.7 Code takes 262,144 tokens; LLM-jp-3.1 8x13B instruct4 takes 4,096.
Can I use Kimi K2.7 Code and LLM-jp-3.1 8x13B instruct4 with the same API key?
Yes. On FastMetal's OpenAI-compatible endpoint, switch by passing "japan-kimi-k2.7-code" or "llm-jp-3.1-8x13b-instruct4" 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 Kimi K2.7 Code and LLM-jp-3.1 8x13B instruct4 from the browser chat and the API. No monthly fee.

More comparisons with Kimi K2.7 Code

More comparisons with LLM-jp-3.1 8x13B instruct4