From d50c93cbfc2acaec55464a00f97033cfddcf44f2 Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Wed, 3 Jan 2024 04:21:05 +0100 Subject: [PATCH] Add Mixtral 2.7, MedicineLLM, MistralLite, and Notux (#1026) * Update models.json Update from Mixtral 8x7b-Dolphin 2.5 to Mixtral 8x7b - Dolphin 2.7 * Update README.md Add Medicine-LLM13B * Update models.json Add Medicine LLM 13B * Update models.json Add MistralLite-7B * Update README.md Add support for MistralLite-7B Add support for Notux 8x7B * Update models.json Add support for MistralLite-7b Add support for Notux 8x7B v1 --------- Co-authored-by: Juan Calderon-Perez <835733+gaby@users.noreply.github.com> --- README.md | 5 ++- api/src/serge/data/models.json | 74 ++++++++++++++++++++++++++++------ 2 files changed, 65 insertions(+), 14 deletions(-) diff --git a/README.md b/README.md index e7ea0ff..e9f47c0 100644 --- a/README.md +++ b/README.md @@ -64,11 +64,14 @@ Instructions for setting up Serge on Kubernetes can be found in the [wiki](https | **LLaMA 2** | 7B, 7B-Chat, 7B-Coder, 13B, 13B-Chat, 70B, 70B-Chat, 70B-OASST | | **Med42** | 70B | | **Medalpaca** | 13B | +| **Medicine-LLM** | 13B | | **Meditron** | 7B, 7B-Chat, 70B | | **Mistral** | 7B-V0.1, 7B-Instruct-v0.2, 7B-OpenOrca | -| **Mixtral** | 8x7B-v0.1, 8x7B-Dophin-2.5, 8x7B-Instruct-v0.1 | +| **MistralLite** | 7B | +| **Mixtral** | 8x7B-v0.1, 8x7B-Dophin-2.7, 8x7B-Instruct-v0.1 | | **Neural-Chat** | 7B-v3.3 | | **Notus** | 7B-v1 | +| **Notux** | 8x7b-v1 | | **OpenChat** | 7B-v3.5-1012 | | **OpenLLaMA** | 3B-v2, 7B-v2, 13B-v2 | | **Orca 2** | 7B, 13B | diff --git a/api/src/serge/data/models.json b/api/src/serge/data/models.json index 15a9dfd..9dde5f0 100644 --- a/api/src/serge/data/models.json +++ b/api/src/serge/data/models.json @@ -319,6 +319,22 @@ ] } ] + }, + { + "name": "medicine-LLM", + "models": [ + { + "name": "Medicine LLM 13B", + "repo": "TheBloke/medicine-LLM-13B-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "medicine-llm-13b.Q4_K_M.gguf", + "disk_space": 7865963456.0 + } + ] + } + ] }, { "name": "Meditron", @@ -371,18 +387,7 @@ "disk_space": 4368438912.0 } ] - }, - { - "name": "Mixtral-8X7B-Dolphin-2_5", - "repo": "TheBloke/dolphin-2.5-mixtral-8x7b-GGUF", - "files": [ - { - "name": "q4_K_M", - "filename": "dolphin-2.5-mixtral-8x7b.Q4_K_M.gguf", - "disk_space": 26441544544.0 - } - ] - }, + }, { "name": "Mistral-7B-Instruct-v0_2", "repo": "TheBloke/Mistral-7B-Instruct-v0.2-GGUF", @@ -407,6 +412,22 @@ } ] }, + { + "name": "MistralLite", + "models": [ + { + "name": "MistralLite-7B", + "repo": "TheBloke/MistralLite-7B-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "mistrallite.Q4_K_M.gguf", + "disk_space": 4368456032.0 + } + ] + } + ] + }, { "name": "Mixtral", "models": [ @@ -421,6 +442,17 @@ } ] }, + { + "name": "Mixtral-8X7B-Dolphin-2_7", + "repo": "TheBloke/dolphin-2.7-mixtral-8x7b-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "dolphin-2.7-mixtral-8x7b.Q4_K_M.gguf", + "disk_space": 26441544512.0 + } + ] + }, { "name": "Mixtral-8X7B-Instruct-v0_1", "repo": "TheBloke/Mixtral-8x7B-Instruct-v0.1-GGUF", @@ -465,7 +497,23 @@ ] } ] - }, + }, + { + "name": "Notux-8X7B", + "models": [ + { + "name": "Notux-8X7B-v1", + "repo": "TheBloke/notux-8x7b-v1-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "notux-8x7b-v1.Q4_K_S.gguf", + "disk_space": 26442469696.0 + } + ] + } + ] + }, { "name": "OpenChat", "models": [