update llama model related config (#392)

Author: key4ng

config/models/meta/Llama-2-13b-chat-hf.yaml

@@ -0,0 +1,16 @@
+# This model is gated on HuggingFace
+apiVersion: ome.io/v1beta1
+kind: ClusterBaseModel
+metadata:
+  name: llama-2-13b-chat-hf
+spec:
+  modelCapabilities:
+    - TEXT_TO_TEXT
+  vendor: meta
+  disabled: false
+  version: "1.0.0"
+  displayName: meta.llama-2-13b-chat-hf
+  storage:
+    storageUri: hf://meta-llama/Llama-2-13b-chat-hf
+    path: /raid/models/meta/llama-2-13b-chat-hf
+    key: "hf-token"

config/models/meta/Llama-2-13b-hf.yaml

@@ -2,15 +2,15 @@
 apiVersion: ome.io/v1beta1
 kind: ClusterBaseModel
 metadata:
-  name: llama-2-13b
+  name: llama-2-13b-hf
 spec:
   modelCapabilities:
     - TEXT_TO_TEXT
   vendor: meta
   disabled: false
   version: "1.0.0"
-  displayName: meta.llama-2-13b
+  displayName: meta.llama-2-13b-hf
   storage:
     storageUri: hf://meta-llama/Llama-2-13b-hf
-    path: /raid/models/meta/llama-2-13b
+    path: /raid/models/meta/llama-2-13b-hf
     key: "hf-token"

config/models/meta/Llama-2-70b-chat-hf.yaml

@@ -0,0 +1,16 @@
+# This model is gated on HuggingFace
+apiVersion: ome.io/v1beta1
+kind: ClusterBaseModel
+metadata:
+  name: llama-2-70b-chat-hf
+spec:
+  modelCapabilities:
+    - TEXT_TO_TEXT
+  vendor: meta
+  disabled: false
+  version: "1.0.0"
+  displayName: meta.llama-2-70b-chat-hf
+  storage:
+    storageUri: hf://meta-llama/Llama-2-70b-chat-hf
+    path: /raid/models/meta/llama-2-70b-chat-hf
+    key: "hf-token"

config/models/meta/Llama-2-70b-hf.yaml

@@ -2,15 +2,15 @@
 apiVersion: ome.io/v1beta1
 kind: ClusterBaseModel
 metadata:
-  name: llama-2-70b
+  name: llama-2-70b-hf
 spec:
   modelCapabilities:
     - TEXT_TO_TEXT
   vendor: meta
   disabled: false
   version: "1.0.0"
-  displayName: meta.llama-2-70b
+  displayName: meta.llama-2-70b-hf
   storage:
     storageUri: hf://meta-llama/Llama-2-70b-hf
-    path: /raid/models/meta/llama-2-70b
+    path: /raid/models/meta/llama-2-70b-hf
     key: "hf-token"

config/models/meta/Llama-2-7b-chat-hf.yaml

@@ -0,0 +1,16 @@
+# This model is gated on HuggingFace
+apiVersion: ome.io/v1beta1
+kind: ClusterBaseModel
+metadata:
+  name: llama-2-7b-chat-hf
+spec:
+  modelCapabilities:
+    - TEXT_TO_TEXT
+  vendor: meta
+  disabled: false
+  version: "1.0.0"
+  displayName: meta.llama-2-7b-chat-hf
+  storage:
+    storageUri: hf://meta-llama/Llama-2-7b-chat-hf
+    path: /raid/models/meta/llama-2-7b-chat-hf
+    key: "hf-token"

config/models/meta/Llama-3.1-405B-Instruct-FP8.yaml

@@ -9,6 +9,15 @@ spec:
   displayName: meta.llama-3.1-405b-instruct-fp8
   disabled: false
   version: "1.0.0"
+  modelFormat:
+    name: safetensors
+    version: "1.0.0"
+  modelFramework:
+    name: transformers
+    version: "4.43.0"
+  modelType: llama
+  modelArchitecture: LlamaForCausalLM
   storage:
     storageUri: hf://meta-llama/Llama-3.1-405B-Instruct-FP8
     path: /raid/models/meta/llama-3-1-405b-instruct-fp8
+    key: "hf-token"

config/models/meta/Llama-4-Maverick-17B-128E-Instruct-FP8.yaml

@@ -19,4 +19,5 @@ spec:
   modelArchitecture: Llama4ForConditionalGeneration
   storage:
     storageUri: hf://meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8
-    path: /raid/models/meta/llama-4-maverick-17b-128e-instruct-fp8
+    path: /raid/models/meta/llama-4-maverick-17b-128e-instruct-fp8
+    key: "hf-token"

config/models/meta/Llama-4-Maverick-17B-128E-Instruct.yaml

@@ -19,4 +19,5 @@ spec:
   modelArchitecture: Llama4ForConditionalGeneration
   storage:
     storageUri: hf://meta-llama/Llama-4-Maverick-17B-128E-Instruct
-    path: /raid/models/meta/llama-4-maverick-17b-128e-instruct
+    path: /raid/models/meta/llama-4-maverick-17b-128e-instruct
+    key: "hf-token"

config/models/meta/Llama-4-Scout-17B-16E-Instruct.yaml

@@ -19,4 +19,5 @@ spec:
   modelArchitecture: Llama4ForConditionalGeneration
   storage:
     storageUri: hf://meta-llama/Llama-4-Scout-17B-16E-Instruct
-    path: /raid/models/meta/llama-4-scout-17b-16e-instruct
+    path: /raid/models/meta/llama-4-scout-17b-16e-instruct
+    key: "hf-token"

config/models/meta/Llama-Guard-3-8B.yaml

@@ -0,0 +1,16 @@
+# This model is gated on HuggingFace
+apiVersion: ome.io/v1beta1
+kind: ClusterBaseModel
+metadata:
+  name: llama-guard-3-8b
+spec:
+  modelCapabilities:
+    - TEXT_TO_TEXT
+  vendor: meta
+  disabled: false
+  version: "1.0.0"
+  displayName: meta.llama-guard-3-8b
+  storage:
+    storageUri: hf://meta-llama/Llama-Guard-3-8B
+    path: /raid/models/meta/llama-guard-3-8b
+    key: "hf-token"