Bedrock
99.98%Context256K
Max output256K
p50 latency705 ms
Throughput55.0 t/s
Input price$0.6
Output price$3
Zero retentionNo training
Model directory
Baseer model profileLanguage model
moonshotai/kimi-k2.5
Kimi K2.5
kimi-k2.5 is Kimi's most versatile model to date, featuring a native multimodal architecture that supports both visual and text input, thinking and non-thinking modes, and dialogue and agent tasks.
Input / 1M tokens$0.6
Output / 1M tokens$3
Maximum context262.1K
Inference providers3
Price comparison
USD per one million tokens, using current provider data.
Provider latency
Last-hour p50 in milliseconds; lower is better.
Provider uptime
Reported availability over the last 24 hours; not a contractual SLA.
Inference providers
Compare current pricing, performance, and privacy across every available provider.
Moonshotai
0.00%Context262.1K
Max output262.1K
p50 latency—
Throughput—
Input price$0.6
Output price$3
Zero retentionNo trainingCaching
Novita
99.84%Context262.1K
Max output262.1K
p50 latency1,006 ms
Throughput39.5 t/s
Input price$0.6
Output price$3
No trainingCaching
| Provider | Maximum context | Max output | p50 latency | p50 throughput | 24h uptime | Input / 1M tokens | Output / 1M tokens | Privacy |
|---|---|---|---|---|---|---|---|---|
Bedrock | 256K | 256K | 705 ms | 55.0 t/s | 99.977% | $0.6 | $3 | |
Moonshotai | 262.1K | 262.1K | — | — | 0.000% | $0.6 | $3 | |
Novita | 262.1K | 262.1K | 1,006 ms | 39.5 t/s | 99.836% | $0.6 | $3 |
What can this model do?
Available modalities and capabilities across inference providers.
TextImageAdvanced reasoningVisionTool useAutomatic cachingVideo inputTools
Technical profile
- Model type
- Language model
- Release date
- Jan 26, 2026
- Knowledge cutoff
- 2025-01
- Regions
- —
- API specifications
- V2, V3, V4
- Temperature control
- Supported
Model identifier
moonshotai/kimi-k2.5Model profile updated: Sep 12, 2026, 1:46 PM
Related models
More options from the same creator or model category.