Qwen4 Max API
Qwen4 Max offers a 1M-token context and 131K output for long documents, reasoning and coding. See this page's pricing table for current rates.
Pricing
Prices may change; billing uses the CNY rate and pricing tier applicable when the request starts.
Call the API
Set KIMISEEK_API_KEY to your KimiSeek API key, then call KimiSeek's /v1/chat/completions endpoint with model set to qwen4-max.
curl https://api.kimiseek.app/v1/chat/completions \
-H "Authorization: Bearer ${KIMISEEK_API_KEY:?Set KIMISEEK_API_KEY in this terminal first.}" \
-H "Content-Type: application/json" \
-d '{
"model": "qwen4-max",
"max_completion_tokens": 1024,
"messages": [
{
"role": "user",
"content": "Hello from KimiSeek"
}
]
}'Where it fits
Best for
- Long documents, specifications and repository-scale reading
- Complex, multi-step reasoning and analysis
- Code review, refactoring and developer tooling
Tradeoffs
- Billing is per million tokens, so long inputs and long outputs raise each request's cost
- Rates may change; the live pricing table above sets the billed rate
Model details
FAQ
What are the 1M-token context and 131K output useful for?
They suit long documents, large code changes and multi-step analysis that would otherwise need chunking. Validate the usable context and per-request cost on your own workload before moving production traffic.
How is cached input priced?
For the standard tier, see the Cached / 1M column in the pricing table above for the rate per million cached input tokens.
How do I call this model?
Set KIMISEEK_API_KEY to your KimiSeek API key, then call KimiSeek's /v1/chat/completions endpoint with model set to qwen4-max.
Which API access methods are currently available?
Currently available protocols: OpenAI Chat Completions · /v1/chat/completions · Responses /v1/responses; Anthropic Messages · /v1/messages; Gemini · /v1beta/models/{model}:generateContent.