Kimi K2 0905 is the September update of Kimi K2 0711, a Mixture-of-Experts (MoE) language model from Moonshot AI with 1 trillion parameters and 32 billion active per pass. The long-context window has been expanded to 256k tokens. This release brings improved agentic coding accuracy and generalization across scaffolds, as well as more aesthetic and functional frontend code for web, 3D, and similar tasks. Kimi K2 remains optimized for advanced tool use, reasoning, and code synthesis, excelling in benchmarks like LiveCodeBench, SWE-bench, ZebraLogic, GPQA, Tau2, and AceBench. Its training uses a novel stack with the MuonClip optimizer for stable large-scale MoE training.
$0.30market price $0.60
$0.112
$1.25market price $2.50
Throughput is how fast the model writes (tokens per second, higher is better). Time to first token is how long before the answer starts (lower is better).
Share of requests served without a server error. Client errors (4xx) do not count.
Not enough throughput data
Not enough TTFT data
Works with the OpenAI SDK pointed at our base URL. Every parameter is in the API reference.
Tokens processed per day on NagaAI.
Kimi K2 0905 is the September update of Kimi K2 0711, a Mixture-of-Experts (MoE) language model from Moonshot AI with 1 trillion parameters and 32 billion active per pass. The long-context window has been expanded to 256k tokens.
On NagaAI, Kimi K2 0905 costs $0.30 per 1M input tokens and $1.25 per 1M output tokens. You pay only for what you use, with no subscription.
Kimi K2 0905 reads up to 262.1k tokens in one request, counting the prompt and the answer together.
Yes. Kimi K2 0905 can call your functions as tools and can answer in a JSON schema you give it.
It takes text as input and returns text.
Create an API key, copy the model ID from the top of this page and send requests with the OpenAI or Anthropic SDK pointed at NagaAI. The API section on this page has a ready example.