From 2835bb403188234bd04b9a89e3792a8c3806047b Mon Sep 17 00:00:00 2001 From: Juan Calderon-Perez Date: Sun, 26 Nov 2023 15:12:30 -0500 Subject: [PATCH 1/2] Add support for Vicuna 7B and 13B --- api/src/serge/data/models.json | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) diff --git a/api/src/serge/data/models.json b/api/src/serge/data/models.json index 16dd0c36576..cb8beafd07d 100644 --- a/api/src/serge/data/models.json +++ b/api/src/serge/data/models.json @@ -1,4 +1,31 @@ [ + { + "name": "Vicuna", + "models": [ + { + "name": "Vicuna-7B", + "repo": "TheBloke/vicuna-7B-v1.5-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "vicuna-7b-v1.5.Q4_K_M.gguf", + "disk_space": 4080000000.0 + } + ] + }, + { + "name": "Vicuna-13B", + "repo": "TheBloke/vicuna-13B-v1.5-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "vicuna-13b-v1.5.Q4_K_M.gguf", + "disk_space": 7870000000.0 + } + ] + } + ] + }, { "name": "CodeLLaMA", "models": [ From dcaedcb0d87ece7b041a93b5c4eac83de37b3820 Mon Sep 17 00:00:00 2001 From: Juan Calderon-Perez Date: Sun, 26 Nov 2023 15:14:03 -0500 Subject: [PATCH 2/2] Update README --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index 72012c86588..263a1485020 100644 --- a/README.md +++ b/README.md @@ -60,6 +60,7 @@ Instructions for setting up Serge on Kubernetes can be found in the [wiki](https | **CodeLLaMA** | 7B, 13B | | **LLaMA** | 7B, 13B, 70B | | **Mistral** | 7B-Instruct, 7B-OpenOrca | +| **Vicuna** | 7B-v1.5, 13B-v1.5 | | **Zephyr** | 7B-Alpha, 7B-Beta | Additional weights can be added to the `serge_weights` volume using `docker cp`: