DeepInfra(@DeepInfra):Encoder-decoder is back 😈!!! In DeepSeek V4.1 Flash the first 20 layers build the global KV that the next 20 read from, so prefill costs about half. Rolled out over the last 24h: throughput doubled, and we are approaching 1T tokens/day on OpenRouter. 30% off to celebrate. Enjoy. Cheapest on the market, as always.

DeepInfra

@DeepInfra

Fast ML inference. Run top AI models using a simple API.

参加 February 2023

76 フォロー中 6K ファン

DeepInfra@DeepInfra

2026.09.18 20:41

Encoder-decoder is back 😈!!! In DeepSeek V4.1 Flash the first 20 layers build the global KV that the next 20 read from, so prefill costs about half. Rolled out over the last 24h: throughput doubled, and we are approaching 1T tokens/day on OpenRouter. 30% off to celebrate. Enjoy. Cheapest on the market, as always.