Same open weights, served by 1 provider. Weights: moonshotai/Kimi-K2-Instruct.
| Provider | Their model id | Serving | Input | Output | Cached in | Blended 3:1 | Source | |
|---|---|---|---|---|---|---|---|---|
DeepInfra Deprecated, 2026-09-07 | moonshotai/Kimi-K2-Instruct | standardMIXED: FP8/FP4Reduced context (131k of 256k) | $0.50 | $2.0 | โ | $0.875 | Verified 12h ago ยท Changed 40d ago |
Same weights don't always mean identical outputs: quantization, context limits and serving stacks can change quality, especially for tool calling. Prices are USD per 1M tokens; hyperscaler prices are for the reference region shown.
| DeepInfra mixed: fp8/fp4 | $90/mo |
Cache hits use each provider's cached-input price where it has one. Cache write costs aren't included.