From a37dc83a9ea7880fe91e4e8794c8edf6bf264f64 Mon Sep 17 00:00:00 2001
From: William De Vena <60664495+williamdevena@users.noreply.github.com>
Date: Wed, 28 Feb 2024 21:19:22 +0100
Subject: [PATCH] langchain_anthropic[patch]: Invoke callback prior to yielding
 token (#18274)

## PR title
langchain_anthropic[patch]: Invoke callback prior to yielding

## PR message
- Description: Invoke callback prior to yielding token in _stream and
_astream methods for anthropic.
- Issue: https://github.com/langchain-ai/langchain/issues/16913
- Dependencies: None
- Twitter handle: None
---
 libs/partners/anthropic/langchain_anthropic/llms.py | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/libs/partners/anthropic/langchain_anthropic/llms.py b/libs/partners/anthropic/langchain_anthropic/llms.py
index d5d04a7f3e..82f1b80433 100644
--- a/libs/partners/anthropic/langchain_anthropic/llms.py
+++ b/libs/partners/anthropic/langchain_anthropic/llms.py
@@ -300,9 +300,10 @@ class AnthropicLLM(LLM, _AnthropicCommon):
             prompt=self._wrap_prompt(prompt), stop_sequences=stop, stream=True, **params
         ):
             chunk = GenerationChunk(text=token.completion)
-            yield chunk
+
             if run_manager:
                 run_manager.on_llm_new_token(chunk.text, chunk=chunk)
+            yield chunk
 
     async def _astream(
         self,
@@ -336,9 +337,10 @@ class AnthropicLLM(LLM, _AnthropicCommon):
             **params,
         ):
             chunk = GenerationChunk(text=token.completion)
-            yield chunk
+
             if run_manager:
                 await run_manager.on_llm_new_token(chunk.text, chunk=chunk)
+            yield chunk
 
     def get_num_tokens(self, text: str) -> int:
         """Calculate number of tokens."""