v0.5.3.post1

https://github.com/vllm-project/vllm/releases/tag/v0.5.3.post1

v0.5.1

https://github.com/vllm-project/vllm/releases/tag/v0.5.1

v0.5.0.post1

https://github.com/vllm-project/vllm/releases/tag/v0.5.0.post1

v0.4-custom.2.dev.5

Add max_num_seqs to deployment script

v0.4-custom.2.dev.2

• Allow for more punica kernel shapes, including a LoRA B which goes to an output size of 64
• Other minor improvements (ed41649, b8b6b1e, etc...)

v0.4-custom.2.dev.1

Add extra punica sizes to support bigger vocabs

v0.4-custom.1

• Falcon/Alfred LoRA support
• Read LoRAs params from env vars
• Add SageMaker support for LoRAs
• Add the /loras endpoint to add LoRAs while the server is running

v0.4-canary.3

• Read LoRAs params from env vars

v0.4-canary.2

• Add SageMaker support for LoRAs
• Add the /loras endpoint to add LoRAs while the server is running

v0.4.0

https://github.com/vllm-project/vllm/releases/tag/v0.4.0.post1