Reddit
Telnyx Lists Kimi K3 at $2.70 / $13.50 per 1M Tokens With $0.27 Cached Input and a 1M-Token Context Window
Telnyx published Kimi K3 inference pricing on 2026-07-28: $2.70 per 1M input tokens, $13.50 per 1M output, and $0.27 per 1M cached input, served with the full 1M-token context window for codebase analysis and multi-turn agent sessions. This is the first concrete third-party hosted price point to appear since the model's release, and the 10:1 cached-input discount is the number that changes the economics for agent loops that replay long prompts. The release notes make no throughput claims and offer no head-to-head price comparison against other providers.
↳ Follow the thread