懒猫微服

版本历史记录

懒猫微服官方

DeepSeek V4.1 Flash 1017 DeepInfer 0.1.10

2026.11.02

Add Token statistics API

懒猫微服官方

DeepSeek V4.1 Flash 1017 DeepInfer 0.1.9

2026.10.26

TTF performance improved by 73.4%.

懒猫微服官方

DeepSeek V4.1 Flash 1017 DeepInfer 0.1.8

2026.10.24

Fix tool invocation in the PI agent.

懒猫微服官方

DeepSeek V4.1 Flash 1017 DeepInfer 0.1.7

2026.10.24

Fix the dialog prompt when there is insufficient space.

懒猫微服官方

DeepSeek V4.1 Flash 1017 DeepInfer 0.1.4

2026.10.24

Initial DeepInfer + EXL3 dual-Thor release with 1M context and 8 request slots; eight consecutive warm 32K prefill samples exceeded 1000 TPS.