From d1ce38450f3417475d72a16ca94c99fcbbd1d82d Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Sun, 25 Feb 2024 20:49:43 +0100 Subject: [PATCH 1/2] Add support for BioMistral, Nous-Hermes-2-Mistral and Nous-Hermes-2-Mixtral (#1156) * Update models.json Add support for BioMistral 7B Add support for Nous-Hermes-2-Mistral-7B-DPO Add support for Nous-Hermes-2-Mixtral-8x7B-DPO Add support for Nous-Hermes-2-Mixtral-8x7B-SFT * Update README.md Add support for BioMistral 7B Add support for Nous-Hermes-2-Mistral-7B-DPO Add support for Nous-Hermes-2-Mixtral-8x7B-DPO Add support for Nous-Hermes-2-Mixtral-8x7B-SFT --- README.md | 2 ++ api/src/serge/data/models.json | 54 ++++++++++++++++++++++++++++++++++ 2 files changed, 56 insertions(+) diff --git a/README.md b/README.md index 1b60e746bf..7a463ff1cc 100644 --- a/README.md +++ b/README.md @@ -58,6 +58,7 @@ Instructions for setting up Serge on Kubernetes can be found in the [wiki](https | Category | Models | |:-------------:|:-------| | **Alfred** | 40B-1023 | +| **BioMistral | 7B | | **Code** | 13B, 33B | | **CodeLLaMA** | 7B, 7B-Instruct, 7B-Python, 13B, 13B-Instruct, 13B-Python, 34B, 34B-Instruct, 34B-Python | | **Gemma** | 2B, 7B | @@ -74,6 +75,7 @@ Instructions for setting up Serge on Kubernetes can be found in the [wiki](https | **Neural-Chat** | 7B-v3.3 | | **Notus** | 7B-v1 | | **Notux** | 8x7b-v1 | +| **Nous-Hermes 2** | Mistral-7B-DPO, Mixtral-8x7B-DPO, Mistral-8x7B-SFT | | **OpenChat** | 7B-v3.5-1210 | | **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 d1624c9a77..5da12fa053 100644 --- a/api/src/serge/data/models.json +++ b/api/src/serge/data/models.json @@ -15,6 +15,22 @@ } ] }, + { + "name": "BioMistral", + "models": [ + { + "name": "BioMistral-7B", + "repo": "BioMistral/BioMistral-7B-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "ggml-model-Q4_K_M.gguf", + "disk_space": 4368439424.0 + } + ] + } + ] + }, { "name": "Code", "models": [ @@ -590,6 +606,44 @@ } ] }, + { + "name": "Nous-Hermes-2", + "models": [ + { + "name": "Nous-Hermes-2-Mistral-7B-DPO", + "repo": "NousResearch/Nous-Hermes-2-Mistral-7B-DPO-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "Nous-Hermes-2-Mistral-7B-DPO.Q4_K_M.gguf", + "disk_space": 4368450560.0 + } + ] + }, + { + "name": "Nous-Hermes-2-Mistral-7B-DPO", + "repo": "TheBloke/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "nous-hermes-2-mixtral-8x7b-dpo.Q4_K_M.gguf", + "disk_space": 28446421792.0 + } + ] + }, + { + "name": "Nous-Hermes-2-Mistral-7B-SFT", + "repo": "NousResearch/Nous-Hermes-2-Mixtral-8x7B-SFT-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "Nous-Hermes-2-Mixtral-8x7B-SFT.Q4_K_M.gguf", + "disk_space": 28446421760.0 + } + ] + } + ] + }, { "name": "OpenChat", "models": [ From 26bf8a3c2938eb57eb86dbb3f0a8e560ef827829 Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Sun, 25 Feb 2024 21:57:30 +0100 Subject: [PATCH 2/2] Update Gemma Models (#1158) * Update Gemma Models Update Gemma 2B & 7B Add support for Gemma 2B-Instruct Add support for Gemma 7B-Instruct * Update Gemma Models Update Gemma 2B & 7B Add support for Gemma 2B-Instruct Add support for Gemma 7B-Instruct --- README.md | 2 +- api/src/serge/data/models.json | 30 ++++++++++++++++++++++++++---- 2 files changed, 27 insertions(+), 5 deletions(-) diff --git a/README.md b/README.md index 7a463ff1cc..91320abc83 100644 --- a/README.md +++ b/README.md @@ -61,7 +61,7 @@ Instructions for setting up Serge on Kubernetes can be found in the [wiki](https | **BioMistral | 7B | | **Code** | 13B, 33B | | **CodeLLaMA** | 7B, 7B-Instruct, 7B-Python, 13B, 13B-Instruct, 13B-Python, 34B, 34B-Instruct, 34B-Python | -| **Gemma** | 2B, 7B | +| **Gemma** | 2B, 2B-Instruct, 7B, 7B-Instruct | | **Falcon** | 7B, 7B-Instruct, 40B, 40B-Instruct | | **LLaMA 2** | 7B, 7B-Chat, 7B-Coder, 13B, 13B-Chat, 70B, 70B-Chat, 70B-OASST | | **LLaMA Pro** | 8B, 8B-Instruct | diff --git a/api/src/serge/data/models.json b/api/src/serge/data/models.json index 5da12fa053..32b18416b5 100644 --- a/api/src/serge/data/models.json +++ b/api/src/serge/data/models.json @@ -216,7 +216,7 @@ "models": [ { "name": "Gemma-2B", - "repo": "MaziyarPanahi/gemma-2b-GGUF", + "repo": "brittlewis12/gemma-2b-GGUF", "files": [ { "name": "q4_K_M", @@ -224,10 +224,21 @@ "disk_space": 1495245728.0 } ] - }, + }, + { + "name": "Gemma-2B-Instruct", + "repo": "brittlewis12/gemma-2b-it-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "gemma-2b-it.Q4_K_M.gguf", + "disk_space": 1630263008.0 + } + ] + }, { "name": "Gemma-7B", - "repo": "MaziyarPanahi/gemma-7b-GGUF", + "repo": "brittlewis12/gemma-7b-GGUF", "files": [ { "name": "q4_K_M", @@ -235,7 +246,18 @@ "disk_space": 5127231648.0 } ] - } + }, + { + "name": "Gemma-7B-Instruct", + "repo": "brittlewis12/gemma-7b-it-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "gemma-7b-it.Q4_K_M.gguf", + "disk_space": 5329759200.0 + } + ] + } ] }, {