kimi-k2-6 - ByteSpike

kimi-k2-6 is Moonshot’s long-context flagship — strong on Chinese tasks with a 128K-token context window.

Pricing

Input	Cache read	Output
$0.95 / 1M	$0.16 / 1M	$4.00 / 1M

Cache write is billed at the input rate. See the pricing table for the live rate.

Protocols

Protocol	Path
OpenAI Chat Completions	`POST https://llm.bytespike.ai/v1/chat/completions`

Quickstart

curl https://llm.bytespike.ai/v1/chat/completions \
  -H "Authorization: Bearer $BYTESPIKE_API_KEY" \
  -H "content-type: application/json" \
  -d '{
    "model": "kimi-k2-6",
    "messages": [{ "role": "user", "content": "Hello, ByteSpike." }]
  }'

Capabilities

Capability	Supported
Chat Completions	✅
Streaming (SSE)	✅
Tools / function calling	✅
JSON mode	✅
Vision (image input)	—
Long context	✅ 128K tokens
Context window	128K tokens

When to use

Long-document Chinese-language QA — 128K context for large-document review.
CN-region multi-doc synthesis — Kimi’s strongest dimension.
Long-context workloads — throw long prompts at it.

Next

claude-opus-4-8 — Anthropic flagship alternative
gemini-3-5-flash — 1M-context alternative