Merge branch 'main' into gpu-support

serge-chat · Jan 4, 2024 · 85c9892 · 85c9892
2 parents 0f3358a + 9b554cc
commit 85c9892
Show file tree

Hide file tree

Showing 9 changed files with 572 additions and 305 deletions.
diff --git a/.github/release-drafter.yml b/.github/release-drafter.yml
@@ -70,7 +70,7 @@ autolabeler:
     - '/(fix|bug|missing|correct)/i'
 - label: '🧹 Updates'
   title: 
-    - '/(improve|update|refactor|deprecated|remove|unused|test)/i'
+    - '/(improve|update|migrate|refactor|deprecated|remove|unused|test)/i'
 - label: '🤖 Dependencies'
   title:
     - '/(bump|dependencies)/i'

diff --git a/README.md b/README.md
@@ -57,21 +57,25 @@ Instructions for setting up Serge on Kubernetes can be found in the [wiki](https
 
 | Category      | Models |
 |:-------------:|:-------|
-| **Alfred** | 40B |
+| **Alfred** | 40B-1023 |
 | **Code** | 13B, 33B |
 | **CodeLLaMA** | 7B, 7B-Instruct, 7B-Python, 13B, 13B-Instruct, 13B-Python, 34B, 34B-Instruct, 34B-Python |
 | **Falcon** | 7B, 7B-Instruct, 40B, 40B-Instruct |
 | **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-Dolphin-2.7, 8x7B-Instruct-v0.1 |
 | **Neural-Chat** | 7B-v3.3 | 
 | **Notus** | 7B-v1 |
-| **OpenChat** | 7B-v3.5-1012 |
+| **Notux** | 8x7b-v1 |
+| **OpenChat** | 7B-v3.5-1210 |
 | **OpenLLaMA** | 3B-v2, 7B-v2, 13B-v2 |
 | **Orca 2** | 7B, 13B |
+| **Phi 2** | 2.7B |
 | **Python Code** | 13B, 33B |
 | **PsyMedRP** | 13B-v1, 20B-v1 |
 | **Starling LM** | 7B-Alpha |
@@ -89,10 +93,6 @@ LLaMA will crash if you don't have enough available memory for the model:
 
 Need help? Join our [Discord](https://discord.gg/62Hc6FEYQH)
 
-## ⭐️ Stargazers
-
-<img src="https://starchart.cc/serge-chat/serge.svg" alt="Stargazers over time" style="max-width: 100%">
-
 ## 🧾 License
 
 [Nathan Sarrazin](https://github.com/nsarrazin) and [Contributors](https://github.com/serge-chat/serge/graphs/contributors). `Serge` is free and open-source software licensed under the [MIT License](https://github.com/serge-chat/serge/blob/main/LICENSE-MIT) and [Apache-2.0](https://github.com/serge-chat/serge/blob/main/LICENSE-APACHE).

diff --git a/api/poetry.lock b/api/poetry.lock
diff --git a/api/pyproject.toml b/api/pyproject.toml
@@ -66,7 +66,7 @@ numpy = "^1.25.2"
 langchain = "^0.0.180"
 loguru = "^0.7.2"
 redis = {extras = ["hiredis"], version = "^5.0.1"}
-pytest = "^7.4.3"
+pytest = "^7.4.4"
 
 [tool.ruff]
 # Enable pycodestyle (`E`) and Pyflakes (`F`) codes by default.

diff --git 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": [
@@ -546,7 +594,23 @@
                 ]
             }
         ]
-    },   
+    },
+    {   
+        "name": "Phi",
+        "models": [
+            {
+                "name": "Phi-2",
+                "repo": "TheBloke/phi-2-GGUF",
+                "files": [
+                    {
+                        "name": "q4_K_M",
+                        "filename": "phi-2.Q4_K_M.gguf",
+                        "disk_space": 1789239136.0
+                    }
+                ]
+            }           
+        ]
+    },    
     {   
         "name": "Python-Code",
         "models": [

diff --git a/scripts/serge.env b/scripts/serge.env
@@ -1 +1 @@
-LLAMA_PYTHON_VERSION=0.2.25
+LLAMA_PYTHON_VERSION=0.2.26