Kimi K2 Instruct: Benchmarks, Pricing & Context Window
Kimi K2 Instruct is a language model from MoonshotAI, released in July 2025.
Kimi K2 is a state-of-the-art mixture-of-experts (MoE) language model with 32 billion activated parameters and 1 trillion total parameters. Trained with the MuonClip optimizer, it achieves exceptional performance across frontier knowledge,
Kimi K2 Instruct pricing
Providers
Kimi K2 Instruct starts at $0.500 per million input tokens and $0.500 per million output tokens via Fireworks. See all 2 providers below with their per-token pricing, latency, throughput, and modality support.
| Provider | Input $/M | Output $/M | Max Input | Max Output | Latency s | Throughput | Quant | Input | Output |
|---|---|---|---|---|---|---|---|---|---|
 Fireworks | $0.500 | $0.500 | 200.0K | 200.0K | — | — | — | ||
 Novita | $0.570 | $2.30 | 131.1K | 131.1K | 0.95 | 45 c/s | fp8 |
Kimi K2 Instruct API
API access coming soon
Kimi K2 Instruct will be available through our gateway shortly.
Kimi K2 Instruct examples
Recent arena outputs from Kimi K2 Instruct, picked from the highest-ranked matchups.
Kimi K2 Instruct license
Kimi K2 Instruct is released under the MIT license, which permits commercial use, has 1.0T parameters.
- License
- MIT
- Commercial use allowed
- Parameters
- 1.0T
MIT License - allows commercial use
FAQ
Common questions about Kimi K2 Instruct.