S-LoRA: Serving Thousands of Concurrent LoRA Adapters · HackerPapers discussions

S-LoRA: Serving Thousands of Concurrent LoRA Adapters — mentioned by sanxiyn

Sun, 18 Feb 2024 04:16:56 GMT

You can also batch requests using different LoRAs. See "S-LoRA: Serving Thousands of Concurrent LoRA Adapters". https://arxiv.org/abs/2311.03285

Paper and discussion history. A mention is not an endorsement.

S-LoRA: Serving Thousands of Concurrent LoRA Adapters — HN submission

Thu, 09 Nov 2023 04:53:52 GMT

S-LoRA: Serving Concurrent LoRA Adapters

41 points; 8 discussion replies on HN. Paper and source history. A submission is not an endorsement.

S-LoRA: Serving Thousands of Concurrent LoRA Adapters — mentioned by kcorbitt

Thu, 09 Nov 2023 00:11:49 GMT

Awesome work! Here's a recent paper released yesterday, also focused on efficiently serving many LoRAs simultaneously: https://arxiv.org/abs/2311.03285 Really looking forward to these innovations becoming more widespread -- I expect we're very close to a world where training a LoRA on a one-off task like "review every HN post from the last 3 years and flag any of them that contain informed speculation about the architecture of GPT-4" will be easy, cheap and routine.

Paper and discussion history. A mention is not an endorsement.