Coding
moonshotai
Kimi K2
Kimi-K2Compare public rates, prepare code review prompts and connect models to your application. Each guide explains how to frame a task and verify the result.
Fixed Usage$0.423486per 1M total tokens
Coding / API
Kimi K2 Thinking API
Compare public rates, prepare code review prompts and connect models to your application. Each guide explains how to frame a task and verify the result.
Kimi-K2-ThinkingmoonshotaiIn the public catalog A catalog listing is not a live availability guarantee. Check /v1/models for access with your API key.
01 / YOUR BRIEF
Start with a focused prompt
Review this diff, identify bugs and propose a minimal patch. Add a regression test for each issue. Preserve the public API.
A sample brief — add your own inputs. This page does not generate content or charge your balance.
03 / SPECIFICATIONS
What the CLODEx route supports
- Model ID
Kimi-K2-Thinking- Protocols
- openai
Only parameters declared by the integration catalog are shown. An omitted parameter is not a promise of support.
04 / DEVELOPERS
One endpoint. Your application.
curl --fail-with-body --no-buffer 'https://clodex.xyz/v1/chat/completions' \
-H "Authorization: Bearer $CLODEX_API_KEY" \
-H 'Content-Type: application/json' \
--data-binary @- <<'CLODEX_JSON'
{"messages":[{"content":"Review this diff, identify bugs and propose a minimal patch. Add a regression test for each issue. Preserve the public API.","role":"user"}],"model":"Kimi-K2-Thinking","stream":true}
CLODEX_JSONKeep your key on the server, never in the browser. A generation example uses your balance only when you run it yourself. Check model access and parameters before submitting.
Documentation ↗Before you submit
Share only relevant files and remove secrets. Review changes and run tests before applying them. A model response does not guarantee correct code.
05 / FAQ
Model FAQ
How do I start using Kimi K2 Thinking in CLODEx?
Sign in to CLODEx, check model access for your key and choose the workspace or API. Reading this page does not submit a model request.
How much does a Kimi K2 Thinking request cost?
Public rates appear in the pricing section. Check the unit: total tokens, separate input and output tokens, per image or a parameter-based video quote. Confirm your account rate in the dashboard.
What if a request does not finish?
Keep the request ID and check the request log. For asynchronous generation, check the existing job's status first. If the cause remains unclear, contact support with the ID, never the API key. Resubmitting may create another paid job.