xKiro
KimiPaid accounts

Kimi K2.8 Preview Agent

Kimi K2.8 Preview Agent is Moonshot AI's Kimi K2.8 Preview coding model served through Kimi's agent gateway, with a 256K-token context window, native tool calling and adjustable reasoning effort.

Get your API keyTry it in chat

Model ID

moonshotai/kimi-k2.8-preview-agent
Context window
262K
tokens
Max output
66K
tokens
Input
$1
per 1M tokens
Output
$4
per 1M tokens

Cache read $0.16

Quickstart

Call Kimi K2.8 Preview Agent in one request.

Point the SDK you already use at xKiro and pass the model ID. The same key works for every model in the catalog.

SDK guides
main.py
import osfrom openai import OpenAI client = OpenAI(    api_key=os.environ["XKIRO_API_KEY"],    base_url="https://api.xkiro.com/v1",) res = client.chat.completions.create(    model="",    messages=[        {"role": "user", "content": "Explain HTTP caching in two sentences."},    ],) print(res.choices[0].message.content)
Response · Example

✓ 200 OK

"model": "moonshotai/kimi-k2.8-preview-agent"

HTTP caching stores copies of responses so later requests can be served without contacting the origin server. Headers such as Cache-Control and ETag decide how long a copy stays fresh and how it is revalidated.

14 input · 42 output tokens · $0.00018 at this model’s rates

Create an API key in the dashboard and export it as XKIRO_API_KEY.

Pricing

What Kimi K2.8 Preview Agent costs.

Billed per million tokens at these rates, from your plan’s budget or your wallet.

  • $1

    Input

    per 1M tokens

  • $4

    Output

    per 1M tokens

  • $0.16

    Cache read

    per 1M tokens

Requires an account that has paid — subscribe to a plan or top up your wallet.

On a plan

From $0.034 / $0.14 on a plan

Plan: up to −96%

Usage on a plan comes out of your allowance, so Kimi K2.8 Preview Agent costs from $0.034 / $0.14 per 1M tokens when you use the whole allowance — up to 96% less than pay-as-you-go.

Compare plans →

Billed fairly

  • Requests that fail with a 5xx error are never billed
  • Retrying an identical blocking request replays the first result — no double charge
  • Cancel a stream and pay only for the tokens generated

Specifications

Limits and capabilities.

Context window
262,144 tokens
Max output
65,536 tokens
Input
Text · Image
Output
Text
Provider
Kimi
Model ID
moonshotai/kimi-k2.8-preview-agent
Good for
Coding · Long context

Capabilities

  • Tool useSupported
  • VisionSupported
  • Video inputNot supported
  • ReasoningSupported

Reasoning effort

  1. none
  2. low
  3. medium
  4. high
  5. max

Set it with reasoning_effort in the OpenAI format, or thinking in the Anthropic format.

Reasoning guide

Supported parameters

  • max_tokens
  • temperature
  • top_p
  • stop
  • frequency_penalty
  • presence_penalty
  • seed
  • stream
  • tools
  • tool_choice
  • response_format
  • structured_outputs
  • reasoning
  • include_reasoning

Compare

Similar models.

Closest in price and context window — same provider first, then alternatives from other providers.

Browse all 69 models
Kimi K2.8 Preview Agent compared with similar models, prices per 1M tokens
ModelInput / 1MOutput / 1M
Kimi K2.8 Preview AgentThis modelPaid accountsVision · Reasoning · Tool use262K$1$4$0.16
Kimi K3 AgentPaid accountsVision · Reasoning · Tool use262K$3$15$0.3
Haiku 4.5Vision · Reasoning · Tool use200K$1$5$0.1
Mistral Large 4Reasoning256K$0.68$2.09$0.07
Grok 4.5Vision · Reasoning · Tool use500K$2$6$0.3
Grok 4.6Paid accountsVision · Reasoning · Tool use500K$2$6$0.5
Grok 4.7Vision · Reasoning · Tool use500K$2$6$0.5
modelmoonshotai/kimi-k2.8-preview-agent

Start building with Kimi K2.8 Preview Agent.

Create a key, point your SDK at xKiro and call moonshotai/kimi-k2.8-preview-agent.

Get your API keyRead the docs

No credit card · Cancel anytime · 26 free models, free forever