Home › Models › Kimi K2.6

Kimi K2.6 API

Moonshot's long-context generalist — 256K context at a fraction of flagship cost, through the Bridge OpenAI-compatible endpoint.

ValueBest for: Long-context agent workhorse at a fraction of K3 cost.

Why developers pick it

• 256K context for long documents and agentic loops at a fraction of K3 cost.
• A strong default for agents that need long histories without flagship pricing.
• Cached-input billing keeps repeated context cheap.

Code example

from openai import OpenAI

client = OpenAI(base_url="https://bridge.computeindex.net/v1", api_key="sk-bridge-YOUR_KEY")

completion = client.chat.completions.create(
    model="kimi-k2.6",
    messages=[
        {"role": "system", "content": "You are a research assistant working through long documents."},
        {"role": "user", "content": open("research.pdf.txt").read()},
    ],
)
print(completion.choices[0].message.content)

FAQ

Is this the same model as Moonshot's official? Yes — the same model, served through the Bridge OpenAI-compatible gateway.

K2.6 or K3? K3 is the flagship with the strongest reasoning; K2.6 covers long-context workloads at a fraction of the cost.

Related