feat: update model listing

src/constants.py

@@ -11,7 +11,6 @@ PROMPTS = [(s + f" ({v} output tokens)", v) for (s, v) in PROMPTS]
 MODEL_REPOSITORY_URL = "https://raw.githubusercontent.com/genai-impact/ecologits/refs/heads/main/ecologits/data/models.json"
 
 main_models_openai = [
-    "chatgpt-4o-latest",
     "gpt-3.5-turbo",
     "gpt-4",
     "gpt-4-turbo",
@@ -19,93 +18,70 @@ main_models_openai = [
     "gpt-4o-mini",
     "o1",
     "o1-mini",
-]
-
-main_models_meta = [
-    "meta-llama/Meta-Llama-3.1-8B",
-    "meta-llama/Meta-Llama-3.1-70B",
-    "meta-llama/Meta-Llama-3.1-405B",
-    "meta-llama/Meta-Llama-3-8B",
-    "meta-llama/Meta-Llama-3-70B",
-    "meta-llama/Meta-Llama-3-70B",
-    "meta-llama/Llama-2-7b",
-    "meta-llama/Llama-2-13b",
-    "meta-llama/Llama-2-70b",
-    "meta-llama/CodeLlama-7b-hf",
-    "meta-llama/CodeLlama-13b-hf",
-    "meta-llama/CodeLlama-34b-hf",
-    "meta-llama/CodeLlama-70b-hf",
-]
-
-main_models_msft = [
-    "microsoft/phi-1",
-    "microsoft/phi-1_5",
-    "microsoft/Phi-3-mini-128k-instruct",
-    "microsoft/Phi-3-small-128k-instruct",
-    "microsoft/Phi-3-medium-128k-instruct",
+    "o3-mini",
+    "gpt-4.1-nano",
+    "gpt-4.1-mini",
+    "gpt-4.1",
+    "o4-mini",
+    "gpt-5-nano",
+    "gpt-5-mini",
+    "gpt-5",
 ]
 
 main_models_anthropic = [
-    "claude-2.0",
-    "claude-2.1",
     "claude-3-5-haiku-latest",
     "claude-3-5-sonnet-latest",
     "claude-3-7-sonnet-latest",
-    "claude-3-haiku-20240307",
-    "claude-3-opus-latest",
-    "claude-3-sonnet-20240229",
+    "claude-opus-4-0",
+    "claude-opus-4-1",
+    "claude-sonnet-4-0",
+    "claude-sonnet-4-5"
 ]
 
 main_models_cohere = [
-    "c4ai-aya-expanse-8b",
-    "c4ai-aya-expanse-32b",
-    "command",
-    "command-light",
+    "command-a-03-2025",
     "command-r",
-    "command-r-plus",
+    "command-r-08-2024",
+    "command-r-plus-08-2024",
+    "command-r7b-12-2024"
 ]
 
 main_models_google = [
-    "google/gemma-2-2b",
-    "google/gemma-2-9b",
-    "google/gemma-2-27b",
-    "google/codegemma-2b",
-    "google/codegemma-7b",
-    "gemini-1.0-pro",
-    "gemini-1.5-pro",
-    "gemini-1.5-flash",
+    "gemini-2.0-flash-lite",
     "gemini-2.0-flash",
-]
-
-main_models_databricks = [
-    "databricks/dolly-v1-6b",
-    "databricks/dolly-v2-12b",
-    "databricks/dolly-v2-7b",
-    "databricks/dolly-v2-3b",
-    "databricks/dbrx-base",
+    "gemini-2.5-flash-lite",
+    "gemini-2.5-flash",
+    "gemini-2.5-pro",
+    "gemma-3-1b-it",
+    "gemma-3-4b-it",
+    "gemma-3-12b-it",
+    "gemma-3-27b-it",
+    "gemma-3n-e2b-it",
+    "gemma-3n-e4b-it"
 ]
 
 main_models_mistral = [
-    "mistralai/Mistral-7B-v0.3",
-    "mistralai/Mixtral-8x7B-v0.1",
-    "mistralai/Mixtral-8x22B-v0.1",
-    "mistralai/Codestral-22B-v0.1",
-    "mistralai/Mathstral-7B-v0.1",
+    "codestral-latest",
+    "devstral-medium-latest",
+    "devstral-small-latest",
+    "magistral-medium-latest",
+    "magistral-small-latest",
     "ministral-3b-latest",
     "ministral-8b-latest",
-    "mistral-tiny",
-    "mistral-small",
-    "mistral-medium",
     "mistral-large-latest",
+    "mistral-medium-latest",
+    "mistral-small-latest",
+    "mistral-tiny-latest",
+    "open-mistral-7b",
+    "open-mistral-nemo",
+    "open-mixtral-8x22b",
+    "open-mixtral-8x7b"
 ]
 
 MAIN_MODELS = (
-    main_models_meta
-    + main_models_openai
+    main_models_openai
     + main_models_anthropic
     + main_models_cohere
-    + main_models_msft
     + main_models_mistral
-    + main_models_databricks
     + main_models_google
 )

src/models.py

@@ -1,8 +1,13 @@
-import requests
 import json
+
+import requests
 import pandas as pd
-from src.constants import MODEL_REPOSITORY_URL, MAIN_MODELS
 import streamlit as st
+from ecologits.model_repository import models as model_repository, ArchitectureTypes
+from ecologits.status_messages import ModelArchNotReleasedWarning, ModelArchMultimodalWarning
+from ecologits.utils.range_value import RangeValue
+
+from src.constants import MODEL_REPOSITORY_URL, MAIN_MODELS
 
 
 def clean_models_data(df, with_filter=True):
@@ -70,10 +75,68 @@ def clean_models_data(df, with_filter=True):
     ]
 
 
+PROVIDERS_FORMAT = {
+    "anthropic": "Anthropic",
+    "cohere": "Cohere",
+    "google_genai": "Google",
+    "mistralai": "Mistral AI",
+    "openai": "OpenAI",
+}
+
+
 @st.cache_data
-def load_models(filter_main=True):
-    resp = requests.get(MODEL_REPOSITORY_URL)
-    data = json.loads(resp.text)
-    df = pd.DataFrame(data["models"])
+def load_models(filter_main=True) -> pd.DataFrame:
+    data = []
+    for m in model_repository.list_models():
+        if filter_main and m.name not in MAIN_MODELS:
+            continue    # Ignore "not main" models when filter is enabled
+
+        if m.architecture.type == ArchitectureTypes.DENSE:
+            if isinstance(m.architecture.parameters, RangeValue):
+                total_parameters = dict(m.architecture.parameters)
+            else:
+                total_parameters = m.architecture.parameters
+            active_parameters = total_parameters
+
+        elif m.architecture.type == ArchitectureTypes.MOE:
+            if isinstance(m.architecture.parameters.total, RangeValue):
+                total_parameters = dict(m.architecture.parameters.total)
+            else:
+                total_parameters = m.architecture.parameters.total
+
+            if isinstance(m.architecture.parameters.active, RangeValue):
+                active_parameters = dict(m.architecture.parameters.active)
+            else:
+                active_parameters = m.architecture.parameters.active
+
+        else:
+            continue    # Ignore model
+
+        warning_arch = False
+        warning_multi_modal = False
+        for w in m.warnings:
+            if isinstance(w, ModelArchNotReleasedWarning):
+                warning_arch = True
+            if isinstance(w, ModelArchMultimodalWarning):
+                warning_multi_modal = True
+
+        data.append({
+            "provider": m.provider.value,
+            "provider_clean": PROVIDERS_FORMAT.get(m.provider.value, m.provider.value),
+            "name": m.name,
+            "name_clean": clean_model_name(m.name),
+            "architecture_type": m.architecture.type.value,
+            "total_parameters": total_parameters,
+            "active_parameters": active_parameters,
+            "warning_arch": warning_arch,
+            "warning_multi_modal": warning_multi_modal,
+        })
+
+    return pd.DataFrame(data)
+
 
-    return clean_models_data(df, filter_main)
+def clean_model_name(model_name: str) -> str:
+    model_name = model_name.replace("latest", "")
+    model_name = model_name.replace("-", " ")
+    model_name = model_name.replace("_", " ")
+    return model_name