Add NECTEC

src/helm/config/model_deployments.yaml

@@ -980,7 +980,22 @@ model_deployments:
     max_sequence_length: 2048
     client_spec:
       class_name: "helm.clients.vision_language.huggingface_vlm_client.HuggingFaceVLMClient"
+
+  ## NECTEC
+  - name: huggingface/Pathumma-llm-text-1.0.0
+    model_name: nectec/Pathumma-llm-text-1.0.0
+    tokenizer_name: nectec/Pathumma-llm-text-1.0.0
+    max_sequence_length: 8192
+    client_spec:
+      class_name: "helm.clients.huggingface_client.HuggingFaceClient"
 
+  - name: huggingface/OpenThaiLLM-Prebuilt-7B
+    model_name: nectec/OpenThaiLLM-Prebuilt-7B
+    tokenizer_name: nectec/OpenThaiLLM-Prebuilt-7B
+    max_sequence_length: 4096
+    client_spec:
+      class_name: "helm.clients.huggingface_client.HuggingFaceClient"
+
   ## KAIST AI
   - name: huggingface/prometheus-vision-13b-v1.0-hf
     model_name: kaistai/prometheus-vision-13b-v1.0-hf

src/helm/config/model_metadata.yaml

@@ -2144,6 +2144,27 @@ models:
     tags: [TEXT_MODEL_TAG, FULL_FUNCTIONALITY_TEXT_MODEL_TAG]
 
 
+
+  # NECTEC
+  - name: nectec/Pathumma-llm-text-1.0.0
+    display_name: Pathumma-llm-text-1.0.0 (7B)
+    description: Pathumma-llm-text-1.0.0 (7B) is a instruction model from  OpenThaiLLM-Prebuilt-7B ([blog](https://medium.com/nectec/pathummallm-v-1-0-0-release-6a098ddfe276))
+    creator_organization_name: nectec
+    access: open
+    num_parameters: 7620000000
+    release_date: 2024-10-28
+    tags: [TEXT_MODEL_TAG, PARTIAL_FUNCTIONALITY_TEXT_MODEL_TAG, INSTRUCTION_FOLLOWING_MODEL_TAG]
+
+  - name: nectec/OpenThaiLLM-Prebuilt-7B
+    display_name: OpenThaiLLM-Prebuilt-7B (7B)
+    description: OpenThaiLLM-Prebuilt-7B (7B) is a pretrained Thai large language model with 7 billion parameters based on Qwen2.5-7B.
+    creator_organization_name: nectec
+    access: open
+    num_parameters: 7620000000
+    release_date: 2024-10-28
+    tags: [TEXT_MODEL_TAG, PARTIAL_FUNCTIONALITY_TEXT_MODEL_TAG]
+
+
 
   # Neurips
   - name: neurips/local
@@ -3459,4 +3480,4 @@ models:
     access: open
     num_parameters: 1380000000
     release: 2024-10-21
-    tags: [TEXT_MODEL_TAG]
+    tags: [TEXT_MODEL_TAG]

src/helm/config/tokenizer_configs.yaml

@@ -456,6 +456,19 @@ tokenizer_configs:
     end_of_text_token: "</s>"
     prefix_token: "<s>"
 
+  # Nectec
+  - name: nectec/OpenThaiLLM-Prebuilt-7B
+    tokenizer_spec:
+      class_name: "helm.tokenizers.huggingface_tokenizer.HuggingFaceTokenizer"
+    end_of_text_token: "<|im_end|>"
+    prefix_token: ""
+
+  - name: nectec/Pathumma-llm-text-1.0.0
+    tokenizer_spec:
+      class_name: "helm.tokenizers.huggingface_tokenizer.HuggingFaceTokenizer"
+    end_of_text_token: "<|im_end|>"
+    prefix_token: "<|im_start|>"
+
   # Neurips
   - name: neurips/local
     tokenizer_spec:
@@ -530,15 +543,15 @@ tokenizer_configs:
       class_name: "helm.tokenizers.huggingface_tokenizer.HuggingFaceTokenizer"
       args:
         pretrained_model_name_or_path: Qwen/Qwen2-72B-Instruct
-    end_of_text_token: <|im_end|>"
+    end_of_text_token: "<|im_end|>"
     prefix_token: "<|im_start|>'"
 
   - name: qwen/qwen2.5-7b-instruct
     tokenizer_spec:
       class_name: "helm.tokenizers.huggingface_tokenizer.HuggingFaceTokenizer"
       args:
         pretrained_model_name_or_path: Qwen/Qwen2.5-7B-Instruct
-    end_of_text_token: <|im_end|>"
+    end_of_text_token: "<|im_end|>"
     prefix_token: "<|im_start|>'"
 
   - name: qwen/qwen-vl
@@ -728,4 +741,4 @@ tokenizer_configs:
       args:
         pretrained_model_name_or_path: ibm-granite/granite-3.0-1b-a400m-base
     end_of_text_token: ""
-    prefix_token: ""
+    prefix_token: ""