Fix multiagent web assistant notebook

Author: aymeric-roucher

notebooks/en/multiagent_web_assistant.ipynb

@@ -32,11 +32,6 @@
     "```\n",
     "Let's set up this system. \n",
     "\n",
-<<<<<<< Updated upstream
-=======
-    "⚡️ Our agent will be powered by [meta-llama/Meta-Llama-3.1-70B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-70B-Instruct) using `HfApiModel` class that uses HF's Inference API: the Inference API allows to quickly and easily run any OS model.\n",
-    "\n",
->>>>>>> Stashed changes
     "Run the line below to install the required dependencies:"
    ]
   },
@@ -46,7 +41,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "!pip install markdownify duckduckgo-search smolagents --upgrade -q"
+    "!pip install markdownify duckduckgo-search \"transformers[agents]\" --upgrade -q"
    ]
   },
   {
@@ -109,7 +104,7 @@
     "import requests\n",
     "from markdownify import markdownify as md\n",
     "from requests.exceptions import RequestException\n",
-    "from smolagents import tool\n",
+    "from transformers.agents import tool\n",
     "\n",
     "\n",
     "@tool\n",
@@ -192,7 +187,7 @@
     "Now that we have all the tools `search` and `visit_webpage`, we can use them to create the web agent.\n",
     "\n",
     "Which configuration to choose for this agent?\n",
-    "- Web browsing is a single-timeline task that does not require parallel tool calls, so JSON tool calling works well for that. We thus choose a `ToolCallingAgent`.\n",
+    "- Web browsing is a single-timeline task that does not require parallel tool calls, so JSON tool calling works well for that. We thus choose a `ReactJsonAgent`.\n",
     "- Also, since sometimes web search requires exploring many pages before finding the correct answer, we prefer to increase the number of `max_iterations` to 10."
    ]
   },
@@ -202,19 +197,19 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from smolagents import (\n",
-    "    CodeAgent,\n",
-    "    ToolCallingAgent,\n",
-    "    HfApiModel,\n",
+    "from transformers.agents import (\n",
+    "    ReactCodeAgent,\n",
+    "    ReactJsonAgent,\n",
+    "    HfApiEngine,\n",
     "    ManagedAgent,\n",
     ")\n",
-    "from smolagents.search import DuckDuckGoSearchTool\n",
+    "from transformers.agents.search import DuckDuckGoSearchTool\n",
     "\n",
-    "model = HfApiModel(model)\n",
+    "llm_engine = HfApiEngine(model)\n",
     "\n",
-    "web_agent = ToolCallingAgent(\n",
+    "web_agent = ReactJsonAgent(\n",
     "    tools=[DuckDuckGoSearchTool(), visit_webpage],\n",
-    "    model=model,\n",
+    "    llm_engine=llm_engine,\n",
     "    max_iterations=10,\n",
     ")"
    ]
@@ -245,7 +240,7 @@
    "source": [
     "Finally we create a manager agent, and upon initialization we pass our managed agent to it in its `managed_agents` argument.\n",
     "\n",
-    "Since this agent is the one tasked with the planning and thinking, advanced reasoning will be beneficial, so a `CodeAgent` will be the best choice.\n",
+    "Since this agent is the one tasked with the planning and thinking, advanced reasoning will be beneficial, so a `ReactCodeAgent` will be the best choice.\n",
     "\n",
     "Also, we want to ask a question that involves the current year: so let us add `additional_authorized_imports=[\"time\", \"datetime\"]`"
    ]
@@ -256,9 +251,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "manager_agent = CodeAgent(\n",
+    "manager_agent = ReactCodeAgent(\n",
     "    tools=[],\n",
-    "    model=model,\n",
+    "    llm_engine=llm_engine,\n",
     "    managed_agents=[managed_web_agent],\n",
     "    additional_authorized_imports=[\"time\", \"datetime\"],\n",
     ")"