Kimi K2 - cheapest: Novita $0.57/M input

Moonshot AI's Kimi K2 model, a 1-trillion-parameter mixture-of-experts system, is now available via cloud providers with Novita offering the lowest input price at $0.57 per million tokens. The model requires approximately 600GB of memory in Q4 quantization and supports a 128k context window, making it accessible primarily through API services.

Kimi K2 MoE premier1T total params, MoE. Requires ~600GB for Q4. Cloud-only for most users. - 1000.0B - 128k - other - Jul 2025 Scores Or run it in the cloud Live per-provider pricing, throughput and uptime - refreshed 31 minutes ago via OpenRouter. Click a column to sort. some pricing may be stale - last verified 2026-07-14 | Provider | Type | Input $/M | Output $/M | Cache $/M | Tok/s | Latency | Uptime | Value | |---|---|---|---|---|---|---|---|---| | | API | 14.00 | 14.00 | - | - | - | - | | | Novita | API | 0.57 | 2.30 | - | - | - | 99.32% | best uptime | Default order: throughput among 95%+ uptime providers, then latency; subscriptions last. Sort by any column. Subscription rows show $/mo in the Value column - per-token columns are "-". Affiliate links are marked sponsored / nofollow. Confirm current pricing on the provider's site before committing. Detailed API pricing page + JSON endpoint → /models/kimi-k2/pricing Inference cost over time Data accumulates from the first daily sync - longer ranges populate over time. Prices come from OpenRouter snapshots, not a historical API.