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)`

libs/community/tests/unit_tests/chat_models/konko.py

@@ -14,7 +14,7 @@ def test_konko_chat_test() -> None:
     """Evaluate basic ChatKonko functionality."""
     chat_instance = ChatKonko(max_tokens=10)
     msg = HumanMessage(content="Hi")
-    chat_response = chat_instance([msg])
+    chat_response = chat_instance.invoke([msg])
     assert isinstance(chat_response, BaseMessage)
     assert isinstance(chat_response.content, str)
 
@@ -23,7 +23,7 @@ def test_konko_chat_test_openai() -> None:
     """Evaluate basic ChatKonko functionality."""
     chat_instance = ChatKonko(max_tokens=10, model="meta-llama/llama-2-70b-chat")
     msg = HumanMessage(content="Hi")
-    chat_response = chat_instance([msg])
+    chat_response = chat_instance.invoke([msg])
     assert isinstance(chat_response, BaseMessage)
     assert isinstance(chat_response.content, str)
 
@@ -48,7 +48,7 @@ def test_konko_system_msg_test() -> None:
     chat_instance = ChatKonko(max_tokens=10)
     sys_msg = SystemMessage(content="Initiate user chat.")
     user_msg = HumanMessage(content="Hi there")
-    chat_response = chat_instance([sys_msg, user_msg])
+    chat_response = chat_instance.invoke([sys_msg, user_msg])
     assert isinstance(chat_response, BaseMessage)
     assert isinstance(chat_response.content, str)
 
@@ -92,7 +92,7 @@ def test_konko_streaming_callback_test() -> None:
         verbose=True,
     )
     msg = HumanMessage(content="Hi")
-    chat_response = chat_instance([msg])
+    chat_response = chat_instance.invoke([msg])
     assert callback_instance.llm_streams > 0
     assert isinstance(chat_response, BaseMessage)
 

libs/community/tests/unit_tests/chat_models/test_llama_edge.py

@@ -72,7 +72,7 @@ def test_wasm_chat_without_service_url() -> None:
     messages = [system_message, user_message]
 
     with pytest.raises(ValueError) as e:
-        chat(messages)
+        chat.invoke(messages)
 
     assert "Error code: 503" in str(e)
     assert "reason: The IP address or port of the chat service is incorrect." in str(e)

libs/community/tests/unit_tests/llms/konko.py

@@ -15,7 +15,7 @@ def test_konko_call() -> None:
         temperature=0.2,
         max_tokens=250,
     )
-    output = llm("Say foo:")
+    output = llm.invoke("Say foo:")
 
     assert llm._llm_type == "konko"
     assert isinstance(output, str)

libs/community/tests/unit_tests/llms/test_callbacks.py

@@ -13,7 +13,7 @@ def test_llm_with_callbacks() -> None:
     """Test LLM callbacks."""
     handler = FakeCallbackHandler()
     llm = FakeListLLM(callbacks=[handler], verbose=True, responses=["foo"])
-    output = llm("foo")
+    output = llm.invoke("foo")
     assert output == "foo"
     assert handler.starts == 1
     assert handler.ends == 1
@@ -26,7 +26,7 @@ def test_chat_model_with_v1_callbacks() -> None:
     llm = FakeListChatModel(
         callbacks=[handler], verbose=True, responses=["fake response"]
     )
-    output = llm([HumanMessage(content="foo")])
+    output = llm.invoke([HumanMessage(content="foo")])
     assert output.content == "fake response"
     assert handler.starts == 1
     assert handler.ends == 1
@@ -41,7 +41,7 @@ def test_chat_model_with_v2_callbacks() -> None:
     llm = FakeListChatModel(
         callbacks=[handler], verbose=True, responses=["fake response"]
     )
-    output = llm([HumanMessage(content="foo")])
+    output = llm.invoke([HumanMessage(content="foo")])
     assert output.content == "fake response"
     assert handler.starts == 1
     assert handler.ends == 1

libs/community/tests/unit_tests/llms/test_gradient_ai.py

@@ -64,7 +64,7 @@ def test_gradient_llm_sync(mocker: MockerFixture, setup: dict) -> None:
     assert llm.gradient_workspace_id == _GRADIENT_WORKSPACE_ID
     assert llm.model_id == _MODEL_ID
 
-    response = llm("Say foo:")
+    response = llm.invoke("Say foo:")
     want = "bar"
 
     assert response == want

libs/community/tests/unit_tests/llms/test_ollama.py

@@ -46,7 +46,7 @@ def test_pass_headers_if_provided(monkeypatch: MonkeyPatch) -> None:
 
     monkeypatch.setattr(requests, "post", mock_post)
 
-    llm("Test prompt")
+    llm.invoke("Test prompt")
 
 
 def test_handle_if_headers_not_provided(monkeypatch: MonkeyPatch) -> None:
@@ -65,7 +65,7 @@ def test_handle_if_headers_not_provided(monkeypatch: MonkeyPatch) -> None:
 
     monkeypatch.setattr(requests, "post", mock_post)
 
-    llm("Test prompt")
+    llm.invoke("Test prompt")
 
 
 def test_handle_kwargs_top_level_parameters(monkeypatch: MonkeyPatch) -> None:
@@ -109,7 +109,7 @@ def test_handle_kwargs_top_level_parameters(monkeypatch: MonkeyPatch) -> None:
 
     monkeypatch.setattr(requests, "post", mock_post)
 
-    llm("Test prompt", model="test-model", system="Test system prompt")
+    llm.invoke("Test prompt", model="test-model", system="Test system prompt")
 
 
 def test_handle_kwargs_with_unknown_param(monkeypatch: MonkeyPatch) -> None:
@@ -157,7 +157,7 @@ def test_handle_kwargs_with_unknown_param(monkeypatch: MonkeyPatch) -> None:
 
     monkeypatch.setattr(requests, "post", mock_post)
 
-    llm("Test prompt", unknown="Unknown parameter value", temperature=0.8)
+    llm.invoke("Test prompt", unknown="Unknown parameter value", temperature=0.8)
 
 
 def test_handle_kwargs_with_options(monkeypatch: MonkeyPatch) -> None:
@@ -189,7 +189,7 @@ def test_handle_kwargs_with_options(monkeypatch: MonkeyPatch) -> None:
 
     monkeypatch.setattr(requests, "post", mock_post)
 
-    llm(
+    llm.invoke(
         "Test prompt",
         model="test-another-model",
         options={"unknown_option": "Unknown option value"},