patch: remove usage of llm, chat model __call__ (#20788)

- `llm(prompt)` -> `llm.invoke(prompt)` - `llm(prompt=prompt` -> `llm.invoke(prompt)` (same with `messages=`) - `llm(prompt, callbacks=callbacks)` -> `llm.invoke(prompt, config={"callbacks": callbacks})` - `llm(prompt, **kwargs)` -> `llm.invoke(prompt, **kwargs)`
2025-09-16 06:53:16 +00:00 · 2024-04-24 19:39:23 -04:00
parent 9b7fb381a4
commit 481d3855dc
181 changed files with 395 additions and 403 deletions
--- a/docs/docs/integrations/callbacks/labelstudio.ipynb
+++ b/docs/docs/integrations/callbacks/labelstudio.ipynb
@@ -194,7 +194,7 @@
    "llm = OpenAI(\n",
    "    temperature=0, callbacks=[LabelStudioCallbackHandler(project_name=\"My Project\")]\n",
    ")\n",
-    "print(llm(\"Tell me a joke\"))"
+    "print(llm.invoke(\"Tell me a joke\"))"
   ]
  },
  {
@@ -270,7 +270,7 @@
    "        )\n",
    "    ]\n",
    ")\n",
-    "llm_results = chat_llm(\n",
+    "llm_results = chat_llm.invoke(\n",
    "    [\n",
    "        SystemMessage(content=\"Always use a lot of emojis\"),\n",
    "        HumanMessage(content=\"Tell me a joke\"),\n",
--- a/docs/docs/integrations/callbacks/llmonitor.md
+++ b/docs/docs/integrations/callbacks/llmonitor.md
@@ -107,7 +107,7 @@ User tracking allows you to identify your users, track their cost, conversations
 from langchain_community.callbacks.llmonitor_callback import LLMonitorCallbackHandler, identify

 with identify("user-123"):
-    llm("Tell me a joke")
+    llm.invoke("Tell me a joke")

 with identify("user-456", user_props={"email": "user456@test.com"}):
    agen.run("Who is Leo DiCaprio's girlfriend?")
--- a/docs/docs/integrations/callbacks/promptlayer.ipynb
+++ b/docs/docs/integrations/callbacks/promptlayer.ipynb
@@ -103,7 +103,7 @@
    "    temperature=0,\n",
    "    callbacks=[PromptLayerCallbackHandler(pl_tags=[\"chatopenai\"])],\n",
    ")\n",
-    "llm_results = chat_llm(\n",
+    "llm_results = chat_llm.invoke(\n",
    "    [\n",
    "        HumanMessage(content=\"What comes after 1,2,3 ?\"),\n",
    "        HumanMessage(content=\"Tell me another joke?\"),\n",
@@ -129,10 +129,11 @@
    "from langchain_community.llms import GPT4All\n",
    "\n",
    "model = GPT4All(model=\"./models/gpt4all-model.bin\", n_ctx=512, n_threads=8)\n",
+    "callbacks = [PromptLayerCallbackHandler(pl_tags=[\"langchain\", \"gpt4all\"])]\n",
    "\n",
-    "response = model(\n",
+    "response = model.invoke(\n",
    "    \"Once upon a time, \",\n",
-    "    callbacks=[PromptLayerCallbackHandler(pl_tags=[\"langchain\", \"gpt4all\"])],\n",
+    "    config={\"callbacks\": callbacks},\n",
    ")"
   ]
  },
@@ -181,7 +182,7 @@
    ")\n",
    "\n",
    "example_prompt = promptlayer.prompts.get(\"example\", version=1, langchain=True)\n",
-    "openai_llm(example_prompt.format(product=\"toasters\"))"
+    "openai_llm.invoke(example_prompt.format(product=\"toasters\"))"
   ]
  },
  {
--- a/docs/docs/integrations/callbacks/trubrics.ipynb
+++ b/docs/docs/integrations/callbacks/trubrics.ipynb
@@ -315,7 +315,7 @@
    }
   ],
   "source": [
-    "chat_res = chat_llm(\n",
+    "chat_res = chat_llm.invoke(\n",
    "    [\n",
    "        SystemMessage(content=\"Every answer of yours must be about OpenAI.\"),\n",
    "        HumanMessage(content=\"Tell me a joke\"),\n",
--- a/docs/docs/integrations/chat/alibaba_cloud_pai_eas.ipynb
+++ b/docs/docs/integrations/chat/alibaba_cloud_pai_eas.ipynb
@@ -72,7 +72,7 @@
   "metadata": {},
   "outputs": [],
   "source": [
-    "output = chat([HumanMessage(content=\"write a funny joke\")])\n",
+    "output = chat.invoke([HumanMessage(content=\"write a funny joke\")])\n",
    "print(\"output:\", output)"
   ]
  },
@@ -90,7 +90,7 @@
   "outputs": [],
   "source": [
    "kwargs = {\"temperature\": 0.8, \"top_p\": 0.8, \"top_k\": 5}\n",
-    "output = chat([HumanMessage(content=\"write a funny joke\")], **kwargs)\n",
+    "output = chat.invoke([HumanMessage(content=\"write a funny joke\")], **kwargs)\n",
    "print(\"output:\", output)"
   ]
  },
--- a/docs/docs/integrations/chat/llama_edge.ipynb
+++ b/docs/docs/integrations/chat/llama_edge.ipynb
@@ -62,7 +62,7 @@
    "messages = [system_message, user_message]\n",
    "\n",
    "# chat with wasm-chat service\n",
-    "response = chat(messages)\n",
+    "response = chat.invoke(messages)\n",
    "\n",
    "print(f\"[Bot] {response.content}\")"
   ]
--- a/docs/docs/integrations/chat/zhipuai.ipynb
+++ b/docs/docs/integrations/chat/zhipuai.ipynb
@@ -119,7 +119,7 @@
   "metadata": {},
   "outputs": [],
   "source": [
-    "response = chat(messages)\n",
+    "response = chat.invoke(messages)\n",
    "print(response.content)  # Displays the AI-generated poem"
   ]
  },
--- a/docs/docs/integrations/llms/anyscale.ipynb
+++ b/docs/docs/integrations/llms/anyscale.ipynb
@@ -147,7 +147,7 @@
    "\n",
    "@ray.remote(num_cpus=0.1)\n",
    "def send_query(llm, prompt):\n",
-    "    resp = llm(prompt)\n",
+    "    resp = llm.invoke(prompt)\n",
    "    return resp\n",
    "\n",
    "\n",
--- a/docs/docs/integrations/llms/aphrodite.ipynb
+++ b/docs/docs/integrations/llms/aphrodite.ipynb
@@ -96,7 +96,7 @@
    ")\n",
    "\n",
    "print(\n",
-    "    llm(\n",
+    "    llm.invoke(\n",
    "        '<|system|>Enter RP mode. You are Ayumu \"Osaka\" Kasuga.<|user|>Hey Osaka. Tell me about yourself.<|model|>'\n",
    "    )\n",
    ")"
--- a/docs/docs/integrations/llms/baichuan.ipynb
+++ b/docs/docs/integrations/llms/baichuan.ipynb
@@ -45,7 +45,7 @@
    "# Load the model\n",
    "llm = BaichuanLLM()\n",
    "\n",
-    "res = llm(\"What's your name?\")\n",
+    "res = llm.invoke(\"What's your name?\")\n",
    "print(res)"
   ]
  },
--- a/docs/docs/integrations/llms/baidu_qianfan_endpoint.ipynb
+++ b/docs/docs/integrations/llms/baidu_qianfan_endpoint.ipynb
@@ -80,7 +80,7 @@
    "os.environ[\"QIANFAN_SK\"] = \"your_sk\"\n",
    "\n",
    "llm = QianfanLLMEndpoint(streaming=True)\n",
-    "res = llm(\"hi\")\n",
+    "res = llm.invoke(\"hi\")\n",
    "print(res)"
   ]
  },
@@ -185,7 +185,7 @@
    "    model=\"ERNIE-Bot-turbo\",\n",
    "    endpoint=\"eb-instant\",\n",
    ")\n",
-    "res = llm(\"hi\")"
+    "res = llm.invoke(\"hi\")"
   ]
  },
  {
--- a/docs/docs/integrations/llms/bittensor.ipynb
+++ b/docs/docs/integrations/llms/bittensor.ipynb
@@ -62,7 +62,7 @@
    "    } \"\"\"\n",
    "\n",
    "multi_response_llm = NIBittensorLLM(top_responses=10)\n",
-    "multi_resp = multi_response_llm(\"What is Neural Network Feeding Mechanism?\")\n",
+    "multi_resp = multi_response_llm.invoke(\"What is Neural Network Feeding Mechanism?\")\n",
    "json_multi_resp = json.loads(multi_resp)\n",
    "pprint(json_multi_resp)"
   ]
--- a/docs/docs/integrations/llms/ctransformers.ipynb
+++ b/docs/docs/integrations/llms/ctransformers.ipynb
@@ -62,7 +62,7 @@
   "metadata": {},
   "outputs": [],
   "source": [
-    "print(llm(\"AI is going to\"))"
+    "print(llm.invoke(\"AI is going to\"))"
   ]
  },
  {
@@ -85,7 +85,7 @@
    "    model=\"marella/gpt-2-ggml\", callbacks=[StreamingStdOutCallbackHandler()]\n",
    ")\n",
    "\n",
-    "response = llm(\"AI is going to\")"
+    "response = llm.invoke(\"AI is going to\")"
   ]
  },
  {
--- a/docs/docs/integrations/llms/ctranslate2.ipynb
+++ b/docs/docs/integrations/llms/ctranslate2.ipynb
@@ -97,7 +97,7 @@
   ],
   "source": [
    "print(\n",
-    "    llm(\n",
+    "    llm.invoke(\n",
    "        \"He presented me with plausible evidence for the existence of unicorns: \",\n",
    "        max_length=256,\n",
    "        sampling_topk=50,\n",
--- a/docs/docs/integrations/llms/deepsparse.ipynb
+++ b/docs/docs/integrations/llms/deepsparse.ipynb
@@ -32,7 +32,7 @@
    "    model=\"zoo:nlg/text_generation/codegen_mono-350m/pytorch/huggingface/bigpython_bigquery_thepile/base-none\"\n",
    ")\n",
    "\n",
-    "print(llm(\"def fib():\"))"
+    "print(llm.invoke(\"def fib():\"))"
   ]
  },
  {
--- a/docs/docs/integrations/llms/edenai.ipynb
+++ b/docs/docs/integrations/llms/edenai.ipynb
@@ -203,7 +203,7 @@
    "User: Answer the following yes/no question by reasoning step by step. Can a dog drive a car?\n",
    "Assistant:\n",
    "\"\"\"\n",
-    "print(llm(prompt))"
+    "print(llm.invoke(prompt))"
   ]
  },
  {
--- a/docs/docs/integrations/llms/google_vertex_ai_palm.ipynb
+++ b/docs/docs/integrations/llms/google_vertex_ai_palm.ipynb
@@ -359,7 +359,7 @@
    "}\n",
    "message = HumanMessage(content=[text_message, image_message])\n",
    "\n",
-    "output = llm([message])\n",
+    "output = llm.invoke([message])\n",
    "print(output.content)"
   ]
  },
@@ -432,7 +432,7 @@
    "}\n",
    "message = HumanMessage(content=[text_message, image_message])\n",
    "\n",
-    "output = llm([message])\n",
+    "output = llm.invoke([message])\n",
    "print(output.content)"
   ]
  },
@@ -457,7 +457,7 @@
   "outputs": [],
   "source": [
    "message2 = HumanMessage(content=\"And where the image is taken?\")\n",
-    "output2 = llm([message, output, message2])\n",
+    "output2 = llm.invoke([message, output, message2])\n",
    "print(output2.content)"
   ]
  },
@@ -486,7 +486,7 @@
    "}\n",
    "message = HumanMessage(content=[text_message, image_message])\n",
    "\n",
-    "output = llm([message])\n",
+    "output = llm.invoke([message])\n",
    "print(output.content)"
   ]
  },
--- a/docs/docs/integrations/llms/koboldai.ipynb
+++ b/docs/docs/integrations/llms/koboldai.ipynb
@@ -57,7 +57,9 @@
   },
   "outputs": [],
   "source": [
-    "response = llm(\"### Instruction:\\nWhat is the first book of the bible?\\n### Response:\")"
+    "response = llm.invoke(\n",
+    "    \"### Instruction:\\nWhat is the first book of the bible?\\n### Response:\"\n",
+    ")"
   ]
  }
 ],
--- a/docs/docs/integrations/llms/konko.ipynb
+++ b/docs/docs/integrations/llms/konko.ipynb
@@ -90,7 +90,7 @@
    "llm = Konko(model=\"mistralai/mistral-7b-v0.1\", temperature=0.1, max_tokens=128)\n",
    "\n",
    "input_ = \"\"\"You are a helpful assistant. Explain Big Bang Theory briefly.\"\"\"\n",
-    "print(llm(input_))"
+    "print(llm.invoke(input_))"
   ]
  },
  {
--- a/docs/docs/integrations/llms/llm_caching.ipynb
+++ b/docs/docs/integrations/llms/llm_caching.ipynb
@@ -1020,7 +1020,7 @@
   "source": [
    "%%time\n",
    "\n",
-    "print(llm(\"Why is the Moon always showing the same side?\"))"
+    "print(llm.invoke(\"Why is the Moon always showing the same side?\"))"
   ]
  },
  {
@@ -1044,7 +1044,7 @@
   "source": [
    "%%time\n",
    "\n",
-    "print(llm(\"Why is the Moon always showing the same side?\"))"
+    "print(llm.invoke(\"Why is the Moon always showing the same side?\"))"
   ]
  },
  {
@@ -1109,7 +1109,7 @@
   "source": [
    "%%time\n",
    "\n",
-    "print(llm(\"Why is the Moon always showing the same side?\"))"
+    "print(llm.invoke(\"Why is the Moon always showing the same side?\"))"
   ]
  },
  {
@@ -1133,7 +1133,7 @@
   "source": [
    "%%time\n",
    "\n",
-    "print(llm(\"How come we always see one face of the moon?\"))"
+    "print(llm.invoke(\"How come we always see one face of the moon?\"))"
   ]
  },
  {
@@ -1238,7 +1238,7 @@
   "source": [
    "%%time\n",
    "\n",
-    "print(llm(\"Is a true fakery the same as a fake truth?\"))"
+    "print(llm.invoke(\"Is a true fakery the same as a fake truth?\"))"
   ]
  },
  {
@@ -1262,7 +1262,7 @@
   "source": [
    "%%time\n",
    "\n",
-    "print(llm(\"Is a true fakery the same as a fake truth?\"))"
+    "print(llm.invoke(\"Is a true fakery the same as a fake truth?\"))"
   ]
  },
  {
@@ -1327,7 +1327,7 @@
   "source": [
    "%%time\n",
    "\n",
-    "print(llm(\"Are there truths that are false?\"))"
+    "print(llm.invoke(\"Are there truths that are false?\"))"
   ]
  },
  {
@@ -1351,7 +1351,7 @@
   "source": [
    "%%time\n",
    "\n",
-    "print(llm(\"Is is possible that something false can be also true?\"))"
+    "print(llm.invoke(\"Is is possible that something false can be also true?\"))"
   ]
  },
  {
--- a/docs/docs/integrations/llms/predibase.ipynb
+++ b/docs/docs/integrations/llms/predibase.ipynb
@@ -96,7 +96,7 @@
   "metadata": {},
   "outputs": [],
   "source": [
-    "response = model(\"Can you recommend me a nice dry wine?\")\n",
+    "response = model.invoke(\"Can you recommend me a nice dry wine?\")\n",
    "print(response)"
   ]
  },
@@ -269,7 +269,7 @@
   "metadata": {},
   "outputs": [],
   "source": [
-    "# response = model(\"Can you help categorize the following emails into positive, negative, and neutral?\")"
+    "# response = model.invoke(\"Can you help categorize the following emails into positive, negative, and neutral?\")"
   ]
  }
 ],
--- a/docs/docs/integrations/llms/replicate.ipynb
+++ b/docs/docs/integrations/llms/replicate.ipynb
@@ -323,7 +323,7 @@
    "User: Answer the following yes/no question by reasoning step by step. Can a dog drive a car?\n",
    "Assistant:\n",
    "\"\"\"\n",
-    "_ = llm(prompt)"
+    "_ = llm.invoke(prompt)"
   ]
  },
  {
@@ -376,13 +376,13 @@
    "Assistant:\n",
    "\"\"\"\n",
    "start_time = time.perf_counter()\n",
-    "raw_output = llm(prompt)  # raw output, no stop\n",
+    "raw_output = llm.invoke(prompt)  # raw output, no stop\n",
    "end_time = time.perf_counter()\n",
    "print(f\"Raw output:\\n {raw_output}\")\n",
    "print(f\"Raw output runtime: {end_time - start_time} seconds\")\n",
    "\n",
    "start_time = time.perf_counter()\n",
-    "stopped_output = llm(prompt, stop=[\"\\n\\n\"])  # stop on double newlines\n",
+    "stopped_output = llm.invoke(prompt, stop=[\"\\n\\n\"])  # stop on double newlines\n",
    "end_time = time.perf_counter()\n",
    "print(f\"Stopped output:\\n {stopped_output}\")\n",
    "print(f\"Stopped output runtime: {end_time - start_time} seconds\")"
--- a/docs/docs/integrations/llms/sparkllm.ipynb
+++ b/docs/docs/integrations/llms/sparkllm.ipynb
@@ -65,7 +65,7 @@
    "# Load the model\n",
    "llm = SparkLLM()\n",
    "\n",
-    "res = llm(\"What's your name?\")\n",
+    "res = llm.invoke(\"What's your name?\")\n",
    "print(res)"
   ]
  },
--- a/docs/docs/integrations/providers/ctransformers.mdx
+++ b/docs/docs/integrations/providers/ctransformers.mdx
@@ -23,7 +23,7 @@ It provides a unified interface for all models:
 ```python
 llm = CTransformers(model='/path/to/ggml-gpt-2.bin', model_type='gpt2')

-print(llm('AI is going to'))
+print(llm.invoke('AI is going to'))
 ```

 If you are getting `illegal instruction` error, try using `lib='avx'` or `lib='basic'`:
--- a/docs/docs/integrations/providers/deepsparse.mdx
+++ b/docs/docs/integrations/providers/deepsparse.mdx
@@ -22,7 +22,7 @@ It provides a unified interface for all models:
 ```python
 llm = DeepSparse(model='zoo:nlg/text_generation/codegen_mono-350m/pytorch/huggingface/bigpython_bigquery_thepile/base-none')

-print(llm('def fib():'))
+print(llm.invoke('def fib():'))
 ```

 Additional parameters can be passed using the `config` parameter:
--- a/docs/docs/integrations/providers/flyte.mdx
+++ b/docs/docs/integrations/providers/flyte.mdx
@@ -83,7 +83,7 @@ def langchain_llm() -> str:
        temperature=0.2,
        callbacks=[FlyteCallbackHandler()],
    )
-    return llm([HumanMessage(content="Tell me a joke")]).content
+    return llm.invoke([HumanMessage(content="Tell me a joke")]).content
 ```

 ### Chain
--- a/docs/docs/integrations/providers/gpt4all.mdx
+++ b/docs/docs/integrations/providers/gpt4all.mdx
@@ -27,7 +27,7 @@ from langchain_community.llms import GPT4All
 model = GPT4All(model="./models/mistral-7b-openorca.Q4_0.gguf", n_threads=8)

 # Generate text
-response = model("Once upon a time, ")
+response = model.invoke("Once upon a time, ")
 ```

 You can also customize the generation parameters, such as n_predict, temp, top_p, top_k, and others.
--- a/docs/docs/integrations/providers/helicone.mdx
+++ b/docs/docs/integrations/providers/helicone.mdx
@@ -29,7 +29,7 @@ openai.api_base = "https://oai.hconeai.com/v1"

 llm = OpenAI(temperature=0.9, headers={"Helicone-Cache-Enabled": "true"})
 text = "What is a helicone?"
-print(llm(text))
+print(llm.invoke(text))
 ```

 [Helicone caching docs](https://docs.helicone.ai/advanced-usage/caching)
@@ -47,7 +47,7 @@ llm = OpenAI(temperature=0.9, headers={
        "Helicone-Property-App": "mobile",
      })
 text = "What is a helicone?"
-print(llm(text))
+print(llm.invoke(text))
 ```

 [Helicone property docs](https://docs.helicone.ai/advanced-usage/custom-properties)
--- a/docs/docs/integrations/providers/konko.mdx
+++ b/docs/docs/integrations/providers/konko.mdx
@@ -44,7 +44,7 @@ See a usage [example](/docs/integrations/llms/konko).
  from langchain.llms import Konko
  llm = Konko(max_tokens=800, model='mistralai/Mistral-7B-v0.1')
  prompt = "Generate a Product Description for Apple Iphone 15"
-  response = llm(prompt)
+  response = llm.invoke(prompt)
  ```

 ## Chat Models
--- a/docs/docs/integrations/providers/predibase.md
+++ b/docs/docs/integrations/providers/predibase.md
@@ -23,7 +23,7 @@ model = Predibase(
    predibase_sdk_version=None,  # optional parameter (defaults to the latest Predibase SDK version if omitted)
 )

-response = model("Can you recommend me a nice dry wine?")
+response = model.invoke("Can you recommend me a nice dry wine?")
 print(response)
 ```

@@ -44,7 +44,7 @@ model = Predibase(
    adapter_version=1,
 )

-response = model("Can you recommend me a nice dry wine?")
+response = model.invoke("Can you recommend me a nice dry wine?")
 print(response)
 ```

@@ -64,6 +64,6 @@ model = Predibase(
    adapter_id="predibase/e2e_nlg",
 )

-response = model("Can you recommend me a nice dry wine?")
+response = model.invoke("Can you recommend me a nice dry wine?")
 print(response)
 ```
--- a/docs/docs/integrations/providers/rwkv.mdx
+++ b/docs/docs/integrations/providers/rwkv.mdx
@@ -44,7 +44,7 @@ def generate_prompt(instruction, input=None):


 model = RWKV(model="./models/RWKV-4-Raven-3B-v7-Eng-20230404-ctx4096.pth", strategy="cpu fp32", tokens_path="./rwkv/20B_tokenizer.json")
-response = model(generate_prompt("Once upon a time, "))
+response = model.invoke(generate_prompt("Once upon a time, "))
 ```
 ## Model File

--- a/docs/docs/use_cases/data_generation.ipynb
+++ b/docs/docs/use_cases/data_generation.ipynb
@@ -545,7 +545,7 @@
    ")\n",
    "\n",
    "_input = prompt.format_prompt(text=dataset[0][\"text\"])\n",
-    "output = llm(_input.to_string())\n",
+    "output = llm.invoke(_input.to_string())\n",
    "\n",
    "parsed = parser.parse(output)\n",
    "parsed"