Merge branch 'wip-v0.4' into typing-dict

Author: mdrxy

.github/workflows/_release.yml

@@ -220,7 +220,7 @@ jobs:
         with:
           python-version: ${{ env.PYTHON_VERSION }}
 
-      - uses: actions/download-artifact@v4
+      - uses: actions/download-artifact@v5
         with:
           name: dist
           path: ${{ inputs.working-directory }}/dist/
@@ -379,7 +379,7 @@ jobs:
         with:
           python-version: ${{ env.PYTHON_VERSION }}
 
-      - uses: actions/download-artifact@v4
+      - uses: actions/download-artifact@v5
         if: startsWith(inputs.working-directory, 'libs/core')
         with:
           name: dist
@@ -447,7 +447,7 @@ jobs:
         with:
           python-version: ${{ env.PYTHON_VERSION }}
 
-      - uses: actions/download-artifact@v4
+      - uses: actions/download-artifact@v5
         with:
           name: dist
           path: ${{ inputs.working-directory }}/dist/
@@ -486,7 +486,7 @@ jobs:
         with:
           python-version: ${{ env.PYTHON_VERSION }}
 
-      - uses: actions/download-artifact@v4
+      - uses: actions/download-artifact@v5
         with:
           name: dist
           path: ${{ inputs.working-directory }}/dist/

.github/workflows/_test_release.yml

@@ -85,7 +85,7 @@ jobs:
     steps:
       - uses: actions/checkout@v4
 
-      - uses: actions/download-artifact@v4
+      - uses: actions/download-artifact@v5
         with:
           name: test-dist
           path: ${{ inputs.working-directory }}/dist/

docs/docs/concepts/async.mdx

@@ -1,4 +1,4 @@
-# Async programming with langchain
+# Async programming with LangChain
 
 :::info Prerequisites
 * [Runnable interface](/docs/concepts/runnables)
@@ -12,7 +12,7 @@ You are expected to be familiar with asynchronous programming in Python before r
 This guide specifically focuses on what you need to know to work with LangChain in an asynchronous context, assuming that you are already familiar with asynchronous programming.
 :::
 
-## Langchain asynchronous APIs
+## LangChain asynchronous APIs
 
 Many LangChain APIs are designed to be asynchronous, allowing you to build efficient and responsive applications.
 

docs/docs/concepts/tools.mdx

@@ -31,7 +31,7 @@ The key attributes that correspond to the tool's **schema**:
 The key methods to execute the function associated with the **tool**:
 
 - **invoke**: Invokes the tool with the given arguments.
-- **ainvoke**: Invokes the tool with the given arguments, asynchronously. Used for [async programming with Langchain](/docs/concepts/async).
+- **ainvoke**: Invokes the tool with the given arguments, asynchronously. Used for [async programming with LangChain](/docs/concepts/async).
 
 ## Create tools using the `@tool` decorator
 

docs/docs/how_to/index.mdx

@@ -34,7 +34,7 @@ These are the core building blocks you can use when building applications.
 [Chat Models](/docs/concepts/chat_models) are newer forms of language models that take messages in and output a message.
 See [supported integrations](/docs/integrations/chat/) for details on getting started with chat models from a specific provider.
 
-- [How to: init any model in one line](/docs/how_to/chat_models_universal_init/)
+- [How to: initialize any model in one line](/docs/how_to/chat_models_universal_init/)
 - [How to: work with local models](/docs/how_to/local_llms)
 - [How to: do function/tool calling](/docs/how_to/tool_calling)
 - [How to: get models to return structured output](/docs/how_to/structured_output)

docs/docs/how_to/query_high_cardinality.ipynb

@@ -15,7 +15,7 @@
    "id": "f2195672-0cab-4967-ba8a-c6544635547d",
    "metadata": {},
    "source": [
-    "# How deal with high cardinality categoricals when doing query analysis\n",
+    "# How to deal with high-cardinality categoricals when doing query analysis\n",
     "\n",
     "You may want to do query analysis to create a filter on a categorical column. One of the difficulties here is that you usually need to specify the EXACT categorical value. The issue is you need to make sure the LLM generates that categorical value exactly. This can be done relatively easy with prompting when there are only a few values that are valid. When there are a high number of valid values then it becomes more difficult, as those values may not fit in the LLM context, or (if they do) there may be too many for the LLM to properly attend to.\n",
     "\n",

docs/docs/how_to/structured_output.ipynb

@@ -614,6 +614,7 @@
     "    HumanMessage(\"Now about caterpillars\", name=\"example_user\"),\n",
     "    AIMessage(\n",
     "        \"\",\n",
+    "        name=\"example_assistant\",\n",
     "        tool_calls=[\n",
     "            {\n",
     "                \"name\": \"joke\",\n",
@@ -909,7 +910,7 @@
     "        ),\n",
     "        (\"human\", \"{query}\"),\n",
     "    ]\n",
-    ").partial(schema=People.schema())\n",
+    ").partial(schema=People.model_json_schema())\n",
     "\n",
     "\n",
     "# Custom parser\n",

docs/docs/integrations/chat/openai.ipynb

@@ -447,6 +447,163 @@
     ")"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "c5d9d19d-8ab1-4d9d-b3a0-56ee4e89c528",
+   "metadata": {},
+   "source": [
+    "### Custom tools\n",
+    "\n",
+    ":::info Requires ``langchain-openai>=0.3.29``\n",
+    "\n",
+    ":::\n",
+    "\n",
+    "[Custom tools](https://platform.openai.com/docs/guides/function-calling#custom-tools) support tools with arbitrary string inputs. They can be particularly useful when you expect your string arguments to be long or complex."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "a47c809b-852f-46bd-8b9e-d9534c17213d",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "================================\u001b[1m Human Message \u001b[0m=================================\n",
+      "\n",
+      "Use the tool to calculate 3^3.\n",
+      "==================================\u001b[1m Ai Message \u001b[0m==================================\n",
+      "\n",
+      "[{'id': 'rs_6894ff5747c0819d9b02fc5645b0be9c000169fd9fb68d99', 'summary': [], 'type': 'reasoning'}, {'call_id': 'call_7SYwMSQPbbEqFcKlKOpXeEux', 'input': 'print(3**3)', 'name': 'execute_code', 'type': 'custom_tool_call', 'id': 'ctc_6894ff5b9f54819d8155a63638d34103000169fd9fb68d99', 'status': 'completed'}]\n",
+      "Tool Calls:\n",
+      "  execute_code (call_7SYwMSQPbbEqFcKlKOpXeEux)\n",
+      " Call ID: call_7SYwMSQPbbEqFcKlKOpXeEux\n",
+      "  Args:\n",
+      "    __arg1: print(3**3)\n",
+      "=================================\u001b[1m Tool Message \u001b[0m=================================\n",
+      "Name: execute_code\n",
+      "\n",
+      "[{'type': 'custom_tool_call_output', 'output': '27'}]\n",
+      "==================================\u001b[1m Ai Message \u001b[0m==================================\n",
+      "\n",
+      "[{'type': 'text', 'text': '27', 'annotations': [], 'id': 'msg_6894ff5db3b8819d9159b3a370a25843000169fd9fb68d99'}]\n"
+     ]
+    }
+   ],
+   "source": [
+    "from langchain_openai import ChatOpenAI, custom_tool\n",
+    "from langgraph.prebuilt import create_react_agent\n",
+    "\n",
+    "\n",
+    "@custom_tool\n",
+    "def execute_code(code: str) -> str:\n",
+    "    \"\"\"Execute python code.\"\"\"\n",
+    "    return \"27\"\n",
+    "\n",
+    "\n",
+    "llm = ChatOpenAI(model=\"gpt-5\", output_version=\"responses/v1\")\n",
+    "\n",
+    "agent = create_react_agent(llm, [execute_code])\n",
+    "\n",
+    "input_message = {\"role\": \"user\", \"content\": \"Use the tool to calculate 3^3.\"}\n",
+    "for step in agent.stream(\n",
+    "    {\"messages\": [input_message]},\n",
+    "    stream_mode=\"values\",\n",
+    "):\n",
+    "    step[\"messages\"][-1].pretty_print()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5ef93be6-6d4c-4eea-acfd-248774074082",
+   "metadata": {},
+   "source": [
+    "<details>\n",
+    "<summary>Context-free grammars</summary>\n",
+    "\n",
+    "OpenAI supports the specification of a [context-free grammar](https://platform.openai.com/docs/guides/function-calling#context-free-grammars) for custom tool inputs in `lark` or `regex` format. See [OpenAI docs](https://platform.openai.com/docs/guides/function-calling#context-free-grammars) for details. The `format` parameter can be passed into `@custom_tool` as shown below:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "2ae04586-be33-49c6-8947-7867801d868f",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "================================\u001b[1m Human Message \u001b[0m=================================\n",
+      "\n",
+      "Use the tool to calculate 3^3.\n",
+      "==================================\u001b[1m Ai Message \u001b[0m==================================\n",
+      "\n",
+      "[{'id': 'rs_689500828a8481a297ff0f98e328689c0681550c89797f43', 'summary': [], 'type': 'reasoning'}, {'call_id': 'call_jzH01RVhu6EFz7yUrOFXX55s', 'input': '3 * 3 * 3', 'name': 'do_math', 'type': 'custom_tool_call', 'id': 'ctc_6895008d57bc81a2b84d0993517a66b90681550c89797f43', 'status': 'completed'}]\n",
+      "Tool Calls:\n",
+      "  do_math (call_jzH01RVhu6EFz7yUrOFXX55s)\n",
+      " Call ID: call_jzH01RVhu6EFz7yUrOFXX55s\n",
+      "  Args:\n",
+      "    __arg1: 3 * 3 * 3\n",
+      "=================================\u001b[1m Tool Message \u001b[0m=================================\n",
+      "Name: do_math\n",
+      "\n",
+      "[{'type': 'custom_tool_call_output', 'output': '27'}]\n",
+      "==================================\u001b[1m Ai Message \u001b[0m==================================\n",
+      "\n",
+      "[{'type': 'text', 'text': '27', 'annotations': [], 'id': 'msg_6895009776b881a2a25f0be8507d08f20681550c89797f43'}]\n"
+     ]
+    }
+   ],
+   "source": [
+    "from langchain_openai import ChatOpenAI, custom_tool\n",
+    "from langgraph.prebuilt import create_react_agent\n",
+    "\n",
+    "grammar = \"\"\"\n",
+    "start: expr\n",
+    "expr: term (SP ADD SP term)* -> add\n",
+    "| term\n",
+    "term: factor (SP MUL SP factor)* -> mul\n",
+    "| factor\n",
+    "factor: INT\n",
+    "SP: \" \"\n",
+    "ADD: \"+\"\n",
+    "MUL: \"*\"\n",
+    "%import common.INT\n",
+    "\"\"\"\n",
+    "\n",
+    "format_ = {\"type\": \"grammar\", \"syntax\": \"lark\", \"definition\": grammar}\n",
+    "\n",
+    "\n",
+    "# highlight-next-line\n",
+    "@custom_tool(format=format_)\n",
+    "def do_math(input_string: str) -> str:\n",
+    "    \"\"\"Do a mathematical operation.\"\"\"\n",
+    "    return \"27\"\n",
+    "\n",
+    "\n",
+    "llm = ChatOpenAI(model=\"gpt-5\", output_version=\"responses/v1\")\n",
+    "\n",
+    "agent = create_react_agent(llm, [do_math])\n",
+    "\n",
+    "input_message = {\"role\": \"user\", \"content\": \"Use the tool to calculate 3^3.\"}\n",
+    "for step in agent.stream(\n",
+    "    {\"messages\": [input_message]},\n",
+    "    stream_mode=\"values\",\n",
+    "):\n",
+    "    step[\"messages\"][-1].pretty_print()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "c63430c9-c7b0-4e92-a491-3f165dddeb8f",
+   "metadata": {},
+   "source": [
+    "</details>"
+   ]
+  },
   {
    "cell_type": "markdown",
    "id": "84833dd0-17e9-4269-82ed-550639d65751",

docs/docs/integrations/providers/gradientai.mdx

@@ -1,4 +1,4 @@
-# ChatGradient
+# DigitalOcean Gradient
 
 This will help you getting started with DigitalOcean Gradient [chat models](/docs/concepts/chat_models).
 

docs/scripts/packages_yml_get_downloads.py

@@ -1,5 +1,6 @@
 from datetime import datetime, timedelta, timezone
 from pathlib import Path
+import re
 
 import requests
 from ruamel.yaml import YAML
@@ -11,10 +12,18 @@
 
 
 def _get_downloads(p: dict) -> int:
-    url = f"https://pypistats.org/api/packages/{p['name']}/recent?period=month"
-    r = requests.get(url)
-    r.raise_for_status()
-    return r.json()["data"]["last_month"]
+    url = f"https://pepy.tech/badge/{p['name']}/month"
+    svg = requests.get(url, timeout=10).text
+    texts = re.findall(r"<text[^>]*>([^<]+)</text>", svg)
+    latest = texts[-1].strip() if texts else "0"
+
+    # parse "1.2k", "3.4M", "12,345" -> int
+    latest = latest.replace(",", "")
+    if latest.endswith(("k", "K")):
+        return int(float(latest[:-1]) * 1_000)
+    if latest.endswith(("m", "M")):
+        return int(float(latest[:-1]) * 1_000_000)
+    return int(float(latest) if "." in latest else int(latest))
 
 
 current_datetime = datetime.now(timezone.utc)