Skip to content

Model Library

KimiReleased July 11, 2025

Kimi K2

Kimi · Released July 11, 2025

View public research Browse the library

Intelligence score

19

At a glance

Intelligence
19
Speed
37.2tokens/sec
Latency
2.12stime to first token
Input price
$0.58per 1M tokens
Output price
$2.40per 1M tokens
Context window
130Ktokens
Parameters
32billions

Summary

Comparison summary

Kimi K2 scores 19 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight non-reasoning models of similar size (median: 17). Kimi K2 generates output at 37.2 tokens per second (based on the median across providers serving the model), which is at the lower end compared to other open weight non-reasoning models of similar size (median: 58.3 t/s). Kimi K2 costs $0.58 per 1M input tokens (better than average, median: $0.60) and $2.40 per 1M output tokens (better than average, median: $2.40), based on the median across providers serving the model.

Specifications

Technical specifications

Model nameKimi K2
ProviderKimi
Release dateJuly 11, 2025
Intelligence score19
Speed37.2 tokens/sec
Latency (TTFT)2.12s
Input price$0.58 / 1M tokens
Output price$2.40 / 1M tokens
Context window130,000 tokens
Parameters32B
ReasoningNo
Input modalitiestext
Output modalities—
Open weightsYes
LicenseModified MIT License
API availability—
SourceModel Intelligence legacy profile: kimi-k2

Pricing

Token prices

Price per 1M tokens

Kimi K2, as listed in the catalog record

Input $0.58 and output $2.40 per 1M tokens. Missing prices are hatched and marked "Not published".
  1. Input$0.58
  2. Output$2.40

Model Intelligence legacy profile: kimi-k2

FAQ · 20

Frequently asked questions

Kimi K2 was released on July 11, 2025.

Kimi K2 was created by Kimi.

Kimi K2 scores 19 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight non-reasoning models of similar size (median: 17).

Kimi K2 generates output at 37.2 tokens per second (based on the median across providers serving the model), which is at the lower end compared to other open weight non-reasoning models of similar size (median: 58.3 t/s).

Kimi K2 has a time to first token (TTFT) of 2.12s (based on the median across providers serving the model), which is better than average compared to other open weight non-reasoning models of similar size (median: 2.33s).

Kimi K2 costs $0.58 per 1M input tokens (better than average, median: $0.60) and $2.40 per 1M output tokens (better than average, median: $2.40), based on the median across providers serving the model.

Kimi K2 costs $0.58 per 1M input tokens and $2.40 per 1M output tokens (based on the median across providers serving the model). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.61 per 1M tokens. Pricing may vary by provider.

No, Kimi K2 is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.

Kimi K2 supports text input.

Kimi K2 supports text output.

No, Kimi K2 does not support image input. It can only process text.

No, Kimi K2 is not multimodal. It only supports text input.

Kimi K2 has a context window of 130k tokens. This determines how much text and conversation history the model can process in a single request.

Yes, Kimi K2 is open weights. The model weights are publicly available and can be downloaded for self-hosting.

Kimi K2 has 1 trillion parameters (32 billion active).

Kimi K2 is a Mixture of Experts (MoE) model with 1 trillion total parameters, but only 32 billion active parameters are used during inference.

Kimi K2 is released under the Modified MIT License license. This license allows commercial use.

Kimi K2 achieves a score of 19 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.

Yes, Kimi K2 is available via API through 2 providers.

Kimi K2 is available through 2 API providers.