From a03f5dd9f66c693d7c4717384a8f1654ceecef3d Mon Sep 17 00:00:00 2001
From: DiamondPoirier <153198166+DiamondPoirier@users.noreply.github.com>
Date: Wed, 12 Feb 2025 17:52:29 +0800
Subject: [PATCH] =?UTF-8?q?Add=20a=20list=20of=20large=20language=20models?=
 =?UTF-8?q?=20of=20deepseek=20and=20image2text=20models=E2=80=A6=20(#4914)?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

### What problem does this PR solve?

#4870
### Type of change

- [x] New Feature (non-breaking change which adds functionality)
---
 conf/llm_factories.json | 125 ++++++++++++++++++++++++++++++++++++++--
 1 file changed, 120 insertions(+), 5 deletions(-)

diff --git a/conf/llm_factories.json b/conf/llm_factories.json
index d87488c2..4a9bf72a 100644
--- a/conf/llm_factories.json
+++ b/conf/llm_factories.json
@@ -98,6 +98,54 @@
             "tags": "LLM,TEXT EMBEDDING,TEXT RE-RANK,TTS,SPEECH2TEXT,MODERATION",
             "status": "1",
             "llm": [
+                {
+                    "llm_name": "deepseek-r1",
+                    "tags": "LLM,CHAT,64K",
+                    "max_tokens": 65792,
+                    "model_type": "chat"
+                },
+                {
+                    "llm_name": "deepseek-v3",
+                    "tags": "LLM,CHAT,64K",
+                    "max_tokens": 65792,
+                    "model_type": "chat"
+                },
+                {
+                    "llm_name": "deepseek-r1-distill-qwen-1.5b",
+                    "tags": "LLM,CHAT,32K",
+                    "max_tokens": 32768,
+                    "model_type": "chat"
+                },
+                {
+                    "llm_name": "deepseek-r1-distill-qwen-7b",
+                    "tags": "LLM,CHAT,32K",
+                    "max_tokens": 32768,
+                    "model_type": "chat"
+                },
+                {
+                    "llm_name": "deepseek-r1-distill-qwen-14b",
+                    "tags": "LLM,CHAT,32K",
+                    "max_tokens": 32768,
+                    "model_type": "chat"
+                },
+                {
+                    "llm_name": "deepseek-r1-distill-qwen-32b",
+                    "tags": "LLM,CHAT,32K",
+                    "max_tokens": 32768,
+                    "model_type": "chat"
+                },
+                {
+                    "llm_name": "deepseek-r1-distill-llama-8b",
+                    "tags": "LLM,CHAT,32K",
+                    "max_tokens": 32768,
+                    "model_type": "chat"
+                },
+                {
+                    "llm_name": "deepseek-r1-distill-llama-70b",
+                    "tags": "LLM,CHAT,32K",
+                    "max_tokens": 32768,
+                    "model_type": "chat"
+                },
                 {
                     "llm_name": "qwen-long",
                     "tags": "LLM,CHAT,10000K",
@@ -1648,7 +1696,8 @@
                     "tags": "TEXT EMBEDDING",
                     "max_tokens": 8129,
                     "model_type": "embedding"
-                },{
+                },
+                {
                     "llm_name": "BAAI/bge-large-en-v1.5",
                     "tags": "TEXT EMBEDDING",
                     "max_tokens": 512,
@@ -1659,17 +1708,20 @@
                     "tags": "TEXT EMBEDDING",
                     "max_tokens": 512,
                     "model_type": "embedding"
-                },{
+                },
+                {
                     "llm_name": "BAAI/bge-small-en-v1.5",
                     "tags": "TEXT EMBEDDING",
                     "max_tokens": 512,
                     "model_type": "embedding"
-                },{
+                },
+                {
                     "llm_name": "nvidia/embed-qa-4",
                     "tags": "TEXT EMBEDDING",
                     "max_tokens": 512,
                     "model_type": "embedding"
-                },{
+                },
+                {
                     "llm_name": "nvidia/llama-3.2-nv-embedqa-1b-v1",
                     "tags": "TEXT EMBEDDING",
                     "max_tokens": 512,
@@ -1704,7 +1756,8 @@
                     "tags": "TEXT EMBEDDING",
                     "max_tokens": 4096,
                     "model_type": "embedding"
-                },{
+                },
+                {
                     "llm_name": "nvidia/nv-embedqa-e5-v5",
                     "tags": "TEXT EMBEDDING",
                     "max_tokens": 1024,
@@ -1757,6 +1810,68 @@
                     "tags": "TEXT EMBEDDING",
                     "max_tokens": 512,
                     "model_type": "embedding"
+                },
+
+
+                {
+                    "llm_name": "adept/fuyu-8b",
+                    "tags": "IMAGE2TEXT,1K",
+                    "max_tokens": 1024,
+                    "model_type": "image2text"
+                },
+                {
+                    "llm_name": "google/deplot",
+                    "tags": "IMAGE2TEXT,8K",
+                    "max_tokens": 8192,
+                    "model_type": "image2text"
+                },
+                {
+                    "llm_name": "google/paligemma",
+                    "tags": "IMAGE2TEXT,256K",
+                    "max_tokens": 256000,
+                    "model_type": "image2text"
+                },
+                {
+                    "llm_name": "meta/llama-3.2-11b-vision-instruct",
+                    "tags": "IMAGE2TEXT,128K",
+                    "max_tokens": 131072,
+                    "model_type": "image2text"
+                },
+                {
+                    "llm_name": "meta/llama-3.2-90b-vision-instruct",
+                    "tags": "IMAGE2TEXT,128K",
+                    "max_tokens": 131072,
+                    "model_type": "image2text"
+                },
+                {
+                    "llm_name": "microsoft/florence-2",
+                    "tags": "IMAGE2TEXT,1K",
+                    "max_tokens": 1024,
+                    "model_type": "image2text"
+                },
+                {
+                    "llm_name": "microsoft/kosmos-2",
+                    "tags": "IMAGE2TEXT,4K",
+                    "max_tokens": 4096,
+                    "model_type": "image2text"
+                },
+                {
+                    "llm_name": "microsoft/phi-3-vision-128k-instruct",
+                    "tags": "IMAGE2TEXT,128K",
+                    "max_tokens": 131072,
+                    "model_type": "image2text"
+                },
+                {
+                    "llm_name": "microsoft/phi-3.5-vision-instruct",
+                    "tags": "IMAGE2TEXT,128K",
+                    "max_tokens": 131072,
+                    "model_type": "image2text"
+                },
+                {
+                    "llm_name": "nvidia/neva-22b",
+                    "tags": "IMAGE2TEXT,1K",
+                    "max_tokens": 1024,
+                    "model_type": "image2text"
                 }
             ]
         },