Skip to content

Model Library

KimiReleased September 5, 2025

Kimi K2 0905

Kimi · Released September 5, 2025

Intelligence score

24

At a glance

Intelligence
24
Speed
37.4tokens/sec
Latency
1.86stime to first token
Input price
$0.60per 1M tokens
Output price
$2.50per 1M tokens
Context window
260Ktokens
Parameters
32billions

Summary

Comparison summary

Kimi K2 0905 scores 24 (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 0905 generates output at 37.4 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 0905 costs $0.60 per 1M input tokens (better than average, median: $0.60) and $2.50 per 1M output tokens (somewhat higher than average, median: $2.40), based on the median across providers serving the model.

Specifications

Technical specifications

Model nameKimi K2 0905
ProviderKimi
Release dateSeptember 5, 2025
Intelligence score24
Speed37.4 tokens/sec
Latency (TTFT)1.86s
Input price$0.60 / 1M tokens
Output price$2.50 / 1M tokens
Context window260,000 tokens
Parameters32B
ReasoningNo
Input modalitiestext
Output modalities—
Open weightsYes
LicenseModified MIT License
API availability—
SourceModel Intelligence legacy profile: kimi-k2-0905

Pricing

Token prices

Price per 1M tokens

Kimi K2 0905, as listed in the catalog record

Input $0.60 and output $2.50 per 1M tokens. Missing prices are hatched and marked "Not published".
  1. Input$0.60
  2. Output$2.50

Model Intelligence legacy profile: kimi-k2-0905

FAQ · 20

Frequently asked questions

Kimi K2 0905 was released on September 5, 2025.

Kimi K2 0905 was created by Kimi.

Kimi K2 0905 scores 24 (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 0905 generates output at 37.4 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 0905 has a time to first token (TTFT) of 1.86s (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 0905 costs $0.60 per 1M input tokens (better than average, median: $0.60) and $2.50 per 1M output tokens (somewhat higher than average, median: $2.40), based on the median across providers serving the model.

Kimi K2 0905 costs $0.60 per 1M input tokens and $2.50 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 $0.79 per 1M tokens. Pricing may vary by provider.

No, Kimi K2 0905 is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.

Kimi K2 0905 supports text input.

Kimi K2 0905 supports text output.

No, Kimi K2 0905 does not support image input. It can only process text.

No, Kimi K2 0905 is not multimodal. It only supports text input.

Kimi K2 0905 has a context window of 260k tokens. This determines how much text and conversation history the model can process in a single request.

Yes, Kimi K2 0905 is open weights. The model weights are publicly available and can be downloaded for self-hosting.

Kimi K2 0905 has 1 trillion parameters (32 billion active).

Kimi K2 0905 is a Mixture of Experts (MoE) model with 1 trillion total parameters, but only 32 billion active parameters are used during inference.

Kimi K2 0905 is released under the Modified MIT License license. This license allows commercial use.

Kimi K2 0905 achieves a score of 24 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.

Yes, Kimi K2 0905 is available via API through 1 provider.

Kimi K2 0905 is available through 1 API provider.