Deepinfra
99.90%Context216.1K
Max output216.1K
p50 latency885 ms
Throughput35.5 t/s
Input price$0.47
Output price$2
Zero retentionNo trainingCaching
Model directory
Baseer model profileLanguage model
moonshotai/kimi-k2-thinking
Kimi K2 Thinking
Kimi K2 Thinking is an advanced open-source thinking model by Moonshot AI. It can execute up to 200 – 300 sequential tool calls without human interference, reasoning coherently across hundreds of steps to solve complex problems. Built as a thinking agent, it reasons step by step while using tools, achieving state-of-the-art performance on Humanity's Last Exam (HLE), BrowseComp, and other benchmarks, with major gains in reasoning, agentic search, coding, writing, and general capabilities.
Input / 1M tokens$0.47
Output / 1M tokens$2
Maximum context216.1K
Inference providers1
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.
| Provider | Maximum context | Max output | p50 latency | p50 throughput | 24h uptime | Input / 1M tokens | Output / 1M tokens | Privacy |
|---|---|---|---|---|---|---|---|---|
Deepinfra | 216.1K | 216.1K | 885 ms | 35.5 t/s | 99.899% | $0.47 | $2 |
What can this model do?
Available modalities and capabilities across inference providers.
TextAdvanced reasoningTool useAutomatic cachingTools
Technical profile
- Model type
- Language model
- Release date
- Nov 6, 2025
- Knowledge cutoff
- 2024-08
- Regions
- —
- API specifications
- V2, V3, V4
- Temperature control
- Supported
Model identifier
moonshotai/kimi-k2-thinkingModel profile updated: Sep 12, 2026, 1:46 PM
Related models
More options from the same creator or model category.