Model Library
KimiReleased January 27, 2026Kimi K2.5
Kimi · Released January 27, 2026
Intelligence score
29
At a glance
- Intelligence
- 29
- Speed
- 43.3tokens/sec
- Latency
- 3.17stime to first token
- Input price
- $0.60per 1M tokens
- Output price
- $3.00per 1M tokens
- Context window
- 260Ktokens
- Parameters
- 32billions
Summary
Comparison summary
Kimi K2.5 (Non-reasoning) scores 29 (estimated) on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 17). Kimi K2.5 (Non-reasoning) generates output at 43.3 tokens per second (based on Kimi's API), which is at the lower end compared to other open weight non-reasoning models of similar size (median: 58.3 t/s). Kimi K2.5 (Non-reasoning) costs $0.60 per 1M input tokens (better than average, median: $0.60) and $3.00 per 1M output tokens (somewhat higher than average, median: $2.40), based on Kimi's API.
Specifications
Technical specifications
| Model name | Kimi K2.5 |
|---|---|
| Provider | Kimi |
| Release date | January 27, 2026 |
| Intelligence score | 29 |
| Speed | 43.3 tokens/sec |
| Latency (TTFT) | 3.17s |
| Input price | $0.60 / 1M tokens |
| Output price | $3.00 / 1M tokens |
| Context window | 260,000 tokens |
| Parameters | 32B |
| Reasoning | No |
| Input modalities | text, image, video |
| Output modalities | — |
| Open weights | Yes |
| License | Modified MIT License |
| API availability | — |
| Source | Model Intelligence legacy profile: kimi-k2-5-non-reasoning |
Pricing
Token prices
Price per 1M tokens
Kimi K2.5, as listed in the catalog record
- Input$0.60
- Output$3.00
| Token type | Price per 1M |
|---|---|
| Input | $0.60 |
| Output | $3.00 |
Model Intelligence legacy profile: kimi-k2-5-non-reasoning
FAQ · 20
Frequently asked questions
Kimi K2.5 (Non-reasoning) was released on January 27, 2026.
Kimi K2.5 (Non-reasoning) was created by Kimi.
Kimi K2.5 (Non-reasoning) scores 29 (estimated) on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 17).
Kimi K2.5 (Non-reasoning) generates output at 43.3 tokens per second (based on Kimi's API), which is at the lower end compared to other open weight non-reasoning models of similar size (median: 58.3 t/s).
Kimi K2.5 (Non-reasoning) has a time to first token (TTFT) of 3.17s (based on Kimi's API), which is at the higher end compared to other open weight non-reasoning models of similar size (median: 2.33s).
Kimi K2.5 (Non-reasoning) costs $0.60 per 1M input tokens (better than average, median: $0.60) and $3.00 per 1M output tokens (somewhat higher than average, median: $2.40), based on Kimi's API.
Kimi K2.5 (Non-reasoning) costs $0.60 per 1M input tokens and $3.00 per 1M output tokens (based on Kimi's API). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.49 per 1M tokens. Pricing may vary by provider.
No, Kimi K2.5 (Non-reasoning) is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
Kimi K2.5 (Non-reasoning) supports text, image, and video input.
Kimi K2.5 (Non-reasoning) supports text output.
Yes, Kimi K2.5 (Non-reasoning) supports image input and can analyze, describe, and answer questions about images.
Yes, Kimi K2.5 (Non-reasoning) is multimodal. It can process text, image, and video input and generate text output.
Kimi K2.5 (Non-reasoning) 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 K2.5 (Non-reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Kimi K2.5 (Non-reasoning) has 1 trillion parameters (32 billion active).
Kimi K2.5 (Non-reasoning) is a Mixture of Experts (MoE) model with 1 trillion total parameters, but only 32 billion active parameters are used during inference.
Kimi K2.5 (Non-reasoning) is released under the Modified MIT License license. This license allows commercial use.
Kimi K2.5 (Non-reasoning) achieves a score of 29 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Yes, Kimi K2.5 (Non-reasoning) is available via API through 7 providers.
Kimi K2.5 (Non-reasoning) is available through 7 API providers.