Back to models/kimi-k2.6
月之暗面
MoonshotCallable

kimi-k2.6

Great for legal work, research reports, enterprise documents, and long-context summarization.

Model code
kimi-k2.6
TextLong contextCaching

Context

200K tokens

Availability

1/1 available

Reference latency

2.50s

Capabilities

Inferred from the model family and tags; actual calls are authoritative.

Function calling
Supported
Structured output
Supported
Vision
Supported
Image generation
Not supported
Web search
Supported
Code execution
Supported
Streaming
Supported
Caching
Supported
Batch inference
Supported

Pricing

Final cost is determined at settlement.

Input$0.93 / M
Output$3.86 / M
Cache hitHit $0.19 / M
Cache write$0.93 / M
Pricing statusUnified pricing

Pricing comes from the unified pricing config; balance and budget are checked before each call.

Limits & context

Rate-limit fields sync from the console config; no guarantees until synced.

Max context
256K tokens
Max output
32K tokens
RPM
No hard cap · fair use
TPM
Tiered by account & key

API examples

Use Turiloop's unified API entry; read the key from an environment variable.

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.TURILOOP_API_KEY,
  baseURL: "https://api.turing.yun/v1"
});

const completion = await client.chat.completions.create({
  model: "kimi-k2.6",
  messages: [{ role: "user", content: "Hello, Turiloop" }]
});

Related models

Other models from the same provider or family