Merge branch 'Ishaan/supported-models' into zhuangqh/preset-test

.github/workflows/kind-cluster/determine_models.py

@@ -21,7 +21,7 @@ def read_yaml(file_path):
 YAML_PR = read_yaml(supp_models_yaml)
 # Format: {falcon-7b : {model_name:falcon-7b, type:text-generation, version: #, tag: #}}
 MODELS = {model['name']: model for model in YAML_PR['models']}
-KAITO_REPO_URL = "https://github.com/kaito-repo/kaito.git"
+KAITO_REPO_URL = "https://github.com/kaito-project/kaito.git"
 
 def set_multiline_output(name, value):
     with open(os.environ['GITHUB_OUTPUT'], 'a') as fh:

presets/models/supported_models.yaml

@@ -32,15 +32,16 @@ models:
   # Falcon
   - name: falcon-7b
     type: text-generation
-    version: https://huggingface.co/tiiuae/falcon-7b/commit/898df1396f35e447d5fe44e0a3ccaaaa69f30d36
+    version: https://huggingface.co/tiiuae/falcon-7b/commit/ec89142b67d748a1865ea4451372db8313ada0d8
     runtime: tfs
-    tag: 0.0.6
+    tag: 0.0.7
   - name: falcon-7b-instruct
     type: text-generation
-    version: https://huggingface.co/tiiuae/falcon-7b-instruct/commit/cf4b3c42ce2fdfe24f753f0f0d179202fea59c99
+    version: https://huggingface.co/tiiuae/falcon-7b-instruct/commit/8782b5c5d8c9290412416618f36a133653e85285
     runtime: tfs
-    tag: 0.0.6
+    tag: 0.0.7
     # Tag history:
+    # 0.0.7 - Support VLLM runtime
     # 0.0.6 - Add Logging & Metrics Server
     # 0.0.5 - Tuning and Adapters
     # 0.0.4 - Adjust default model params (#310)
@@ -49,15 +50,16 @@ models:
     # 0.0.1 - Initial Release
   - name: falcon-40b
     type: text-generation
-    version: https://huggingface.co/tiiuae/falcon-40b/commit/4a70170c215b36a3cce4b4253f6d0612bb7d4146
+    version: https://huggingface.co/tiiuae/falcon-40b/commit/05ab2ee8d6b593bdbab17d728de5c028a7a94d83
     runtime: tfs
-    tag: 0.0.7
+    tag: 0.0.8
   - name: falcon-40b-instruct
     type: text-generation
     version: https://huggingface.co/tiiuae/falcon-40b-instruct/commit/ecb78d97ac356d098e79f0db222c9ce7c5d9ee5f
     runtime: tfs
-    tag: 0.0.7
+    tag: 0.0.8
     # Tag history for 40b models:
+    # 0.0.8 - Support VLLM runtime
     # 0.0.7 - Add Logging & Metrics Server
     # 0.0.6 - Tuning and Adapters
     # 0.0.5 - Adjust default model params (#310)
@@ -69,15 +71,16 @@ models:
   # Mistral
   - name: mistral-7b
     type: text-generation 
-    version: https://huggingface.co/mistralai/Mistral-7B-v0.3/commit/c882233d224d27b727b3d9299b12a9aab9dda6f7
+    version: https://huggingface.co/mistralai/Mistral-7B-v0.3/commit/d8cadc02ac76bd617a919d50b092e59d2d110aff
     runtime: tfs
-    tag: 0.0.7
+    tag: 0.0.8
   - name: mistral-7b-instruct
     type: text-generation
-    version: https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3/commit/0417f4babd26db0b5ed07c1d0bc85658ab526ea3
+    version: https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3/commit/e0bc86c23ce5aae1db576c8cca6f06f1f73af2db
     runtime: tfs
-    tag: 0.0.7
+    tag: 0.0.8
     # Tag history:
+    # 0.0.8 - Support VLLM runtime
     # 0.0.7 - Add Logging & Metrics Server
     # 0.0.6 - Update model version and Address missing weights files fix
     # 0.0.5 - Tuning and Adapters
@@ -89,10 +92,11 @@ models:
   # Phi-2
   - name: phi-2
     type: text-generation 
-    version: https://huggingface.co/microsoft/phi-2/commit/b10c3eba545ad279e7208ee3a5d644566f001670
+    version: https://huggingface.co/microsoft/phi-2/commit/ef382358ec9e382308935a992d908de099b64c23
     runtime: tfs
-    tag: 0.0.5
+    tag: 0.0.6
     # Tag history:
+    # 0.0.6 - Support VLLM runtime
     # 0.0.5 - Add Logging & Metrics Server
     # 0.0.4 - Tuning and Adapters
     # 0.0.3 - Adjust default model params (#310)
@@ -102,36 +106,49 @@ models:
   # Phi-3
   - name: phi-3-mini-4k-instruct
     type: text-generation 
-    version: https://huggingface.co/microsoft/Phi-3-mini-4k-instruct/commit/d269012bea6fbe38ce7752c8940fea010eea3383
+    version: https://huggingface.co/microsoft/Phi-3-mini-4k-instruct/commit/0a67737cc96d2554230f90338b163bc6380a2a85
     runtime: tfs
-    tag: 0.0.2
+    tag: 0.0.3
     # Tag history:
+    # 0.0.3 - Support VLLM runtime
     # 0.0.2 - Add Logging & Metrics Server
     # 0.0.1 - Initial Release
 
   - name: phi-3-mini-128k-instruct
     type: text-generation 
-    version: https://huggingface.co/microsoft/Phi-3-mini-128k-instruct/commit/5be6479b4bc06a081e8f4c6ece294241ccd32dec
+    version: https://huggingface.co/microsoft/Phi-3-mini-128k-instruct/commit/a90b62ae09941edff87a90ced39ba5807e6b2ade
     runtime: tfs
-    tag: 0.0.2
+    tag: 0.0.3
     # Tag history:
+    # 0.0.3 - Support VLLM runtime
     # 0.0.2 - Add Logging & Metrics Server
     # 0.0.1 - Initial Release
 
   - name: phi-3-medium-4k-instruct
     type: text-generation
-    version: https://huggingface.co/microsoft/Phi-3-medium-4k-instruct/commit/d194e4e74ffad5a5e193e26af25bcfc80c7f1ffc
+    version: https://huggingface.co/microsoft/Phi-3-medium-4k-instruct/commit/ae004ae82eb6eddc32906dfacb1d6dfea8f91996
     runtime: tfs
-    tag: 0.0.2
+    tag: 0.0.3
     # Tag history:
+    # 0.0.3 - Support VLLM runtime
     # 0.0.2 - Add Logging & Metrics Server
     # 0.0.1 - Initial Release
 
   - name: phi-3-medium-128k-instruct
     type: text-generation
-    version: https://huggingface.co/microsoft/Phi-3-medium-128k-instruct/commit/cae1d42b5577398fd1be9f0746052562ae552886
+    version: https://huggingface.co/microsoft/Phi-3-medium-128k-instruct/commit/fa7d2aa4f5ea69b2e36b20d050cdae79c9bfbb3f
     runtime: tfs
-    tag: 0.0.2
+    tag: 0.0.3
     # Tag history:
+    # 0.0.3 - Support VLLM runtime
     # 0.0.2 - Add Logging & Metrics Server
-    # 0.0.1 - Initial Release
+    # 0.0.1 - Initial Release
+
+  - name: phi-3.5-mini-instruct
+    type: text-generation
+    version: https://huggingface.co/microsoft/Phi-3.5-mini-instruct/commit/af0dfb8029e8a74545d0736d30cb6b58d2f0f3f0
+    runtime: tfs
+    tag: 0.0.1
+    # Tag history:
+    # 0.0.1 - New Model! Support VLLM Runtime
+