From 39ed967c80a55ad226f662f7e5123f3c79235920 Mon Sep 17 00:00:00 2001
From: usamimeri_renko <1710269958@qq.com>
Date: Thu, 2 May 2024 01:48:04 +0800
Subject: [PATCH 01/18] make rag configurable

---
 config/config2.example.yaml         |  1 +
 metagpt/configs/embedding_config.py |  4 ++++
 metagpt/rag/schema.py               | 12 +++++++++---
 3 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/config/config2.example.yaml b/config/config2.example.yaml
index 3249f5ae3..6d1148a85 100644
--- a/config/config2.example.yaml
+++ b/config/config2.example.yaml
@@ -18,6 +18,7 @@ embedding:
   model: ""
   api_version: ""
   embed_batch_size: 100
+  dimensions: 
 
 repair_llm_output: true  # when the output is not a valid json, try to repair it
 
diff --git a/metagpt/configs/embedding_config.py b/metagpt/configs/embedding_config.py
index 20de47999..f9b41b9dc 100644
--- a/metagpt/configs/embedding_config.py
+++ b/metagpt/configs/embedding_config.py
@@ -20,11 +20,13 @@ class EmbeddingConfig(YamlModel):
     ---------
     api_type: "openai"
     api_key: "YOU_API_KEY"
+    dimensions: "YOUR_MODEL_DIMENSIONS"
 
     api_type: "azure"
     api_key: "YOU_API_KEY"
     base_url: "YOU_BASE_URL"
     api_version: "YOU_API_VERSION"
+    dimensions: "YOUR_MODEL_DIMENSIONS"
 
     api_type: "gemini"
     api_key: "YOU_API_KEY"
@@ -32,6 +34,7 @@ class EmbeddingConfig(YamlModel):
     api_type: "ollama"
     base_url: "YOU_BASE_URL"
     model: "YOU_MODEL"
+    dimensions: "YOUR_MODEL_DIMENSIONS"
     """
 
     api_type: Optional[EmbeddingType] = None
@@ -41,6 +44,7 @@ class EmbeddingConfig(YamlModel):
 
     model: Optional[str] = None
     embed_batch_size: Optional[int] = None
+    dimensions: Optional[int] = None  # output dimension of embedding model
 
     @field_validator("api_type", mode="before")
     @classmethod
diff --git a/metagpt/rag/schema.py b/metagpt/rag/schema.py
index e7b2e5ce9..bedba164c 100644
--- a/metagpt/rag/schema.py
+++ b/metagpt/rag/schema.py
@@ -12,6 +12,7 @@ from pydantic import BaseModel, ConfigDict, Field, PrivateAttr, model_validator
 
 from metagpt.config2 import config
 from metagpt.configs.embedding_config import EmbeddingType
+from metagpt.logs import logger
 from metagpt.rag.interface import RAGObject
 
 
@@ -34,16 +35,21 @@ class IndexRetrieverConfig(BaseRetrieverConfig):
 class FAISSRetrieverConfig(IndexRetrieverConfig):
     """Config for FAISS-based retrievers."""
 
-    dimensions: int = Field(default=0, description="Dimensionality of the vectors for FAISS index construction.")
+    dimensions: int = Field(
+        default=config.embedding.dimensions, description="Dimensionality of the vectors for FAISS index construction."
+    )
 
     _embedding_type_to_dimensions: ClassVar[dict[EmbeddingType, int]] = {
         EmbeddingType.GEMINI: 768,
-        EmbeddingType.OLLAMA: 4096,
     }
 
     @model_validator(mode="after")
     def check_dimensions(self):
-        if self.dimensions == 0:
+        if self.dimensions is None:
+            if config.embedding.api_type not in self._embedding_type_to_dimensions:
+                logger.info(
+                    f"You didn't set the dimensions in config when using {config.embedding.api_type}, default to 1536"
+                )
             self.dimensions = self._embedding_type_to_dimensions.get(config.embedding.api_type, 1536)
 
         return self

From fa8b35cef4fb4de9db2b0f0b79cfe85718e26775 Mon Sep 17 00:00:00 2001
From: usamimeri_renko <1710269958@qq.com>
Date: Fri, 3 May 2024 16:46:41 +0800
Subject: [PATCH 02/18] add comment

---
 config/config2.example.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/config/config2.example.yaml b/config/config2.example.yaml
index 6d1148a85..e57ec3ee8 100644
--- a/config/config2.example.yaml
+++ b/config/config2.example.yaml
@@ -18,7 +18,7 @@ embedding:
   model: ""
   api_version: ""
   embed_batch_size: 100
-  dimensions: 
+  dimensions: # output dimension of embedding model
 
 repair_llm_output: true  # when the output is not a valid json, try to repair it
 

From d4c0678aafbdd9617d1c0096d6e8e21b613c8abe Mon Sep 17 00:00:00 2001
From: seehi <6580@pm.me>
Date: Mon, 6 May 2024 14:52:46 +0800
Subject: [PATCH 03/18] fix potential NoneType error

---
 metagpt/rag/retrievers/bm25_retriever.py | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/metagpt/rag/retrievers/bm25_retriever.py b/metagpt/rag/retrievers/bm25_retriever.py
index 241820cf4..dc75d87b0 100644
--- a/metagpt/rag/retrievers/bm25_retriever.py
+++ b/metagpt/rag/retrievers/bm25_retriever.py
@@ -40,8 +40,10 @@ class DynamicBM25Retriever(BM25Retriever):
         self._corpus = [self._tokenizer(node.get_content()) for node in self._nodes]
         self.bm25 = BM25Okapi(self._corpus)
 
-        self._index.insert_nodes(nodes, **kwargs)
+        if self._index:
+            self._index.insert_nodes(nodes, **kwargs)
 
     def persist(self, persist_dir: str, **kwargs) -> None:
         """Support persist."""
-        self._index.storage_context.persist(persist_dir)
+        if self._index:
+            self._index.storage_context.persist(persist_dir)

From 53369ee6da1b3b1090bd9399b537ef99f2e5218a Mon Sep 17 00:00:00 2001
From: seehi <6580@pm.me>
Date: Mon, 6 May 2024 15:52:18 +0800
Subject: [PATCH 04/18] update comment in rag_pipeline example

---
 examples/rag_pipeline.py | 32 +++++++++++++++++++-------------
 1 file changed, 19 insertions(+), 13 deletions(-)

diff --git a/examples/rag_pipeline.py b/examples/rag_pipeline.py
index 7dbca35a6..75888e480 100644
--- a/examples/rag_pipeline.py
+++ b/examples/rag_pipeline.py
@@ -18,13 +18,13 @@ from metagpt.rag.schema import (
 )
 from metagpt.utils.exceptions import handle_exception
 
+LLM_TIP = "If you not sure, just answer I don't know."
+
 DOC_PATH = EXAMPLE_DATA_PATH / "rag/writer.txt"
-QUESTION = "What are key qualities to be a good writer?"
+QUESTION = f"What are key qualities to be a good writer? {LLM_TIP}"
 
 TRAVEL_DOC_PATH = EXAMPLE_DATA_PATH / "rag/travel.txt"
-TRAVEL_QUESTION = "What does Bob like?"
-
-LLM_TIP = "If you not sure, just answer I don't know."
+TRAVEL_QUESTION = f"What does Bob like? {LLM_TIP}"
 
 
 class Player(BaseModel):
@@ -40,21 +40,21 @@ class Player(BaseModel):
 
 
 class RAGExample:
-    """Show how to use RAG.
+    """Show how to use RAG."""
 
-    Default engine use LLM Reranker, if the answer from the LLM is incorrect, may encounter `IndexError: list index out of range`.
-    """
-
-    def __init__(self, engine: SimpleEngine = None):
+    def __init__(self, engine: SimpleEngine = None, use_llm_ranker: bool = False):
         self._engine = engine
+        self._use_llm_ranker = use_llm_ranker
 
     @property
     def engine(self):
         if not self._engine:
+            ranker_configs = [LLMRankerConfig()] if self._use_llm_ranker else None
+
             self._engine = SimpleEngine.from_docs(
                 input_files=[DOC_PATH],
                 retriever_configs=[FAISSRetrieverConfig()],
-                ranker_configs=[LLMRankerConfig()],
+                ranker_configs=ranker_configs,
             )
         return self._engine
 
@@ -105,7 +105,7 @@ class RAGExample:
         """
         self._print_title("Add Docs")
 
-        travel_question = f"{TRAVEL_QUESTION}{LLM_TIP}"
+        travel_question = f"{TRAVEL_QUESTION}"
         travel_filepath = TRAVEL_DOC_PATH
 
         logger.info("[Before add docs]")
@@ -240,8 +240,14 @@ class RAGExample:
 
 
 async def main():
-    """RAG pipeline."""
-    e = RAGExample()
+    """RAG pipeline.
+
+    Note:
+    1. If `use_llm_ranker` is True, then will use LLM Reranker to get better result, but it is not always guaranteed that the output will be parseable for reranking,
+       prefer `gpt-4-turbo`, otherwise might encounter `IndexError: list index out of range` or `ValueError: invalid literal for int() with base 10`.
+    """
+    e = RAGExample(use_llm_ranker=False)
+
     await e.run_pipeline()
     await e.add_docs()
     await e.add_objects()

From 553702fa6131b042c0af272de00860a27e506c31 Mon Sep 17 00:00:00 2001
From: usamimeri_renko <1710269958@qq.com>
Date: Tue, 7 May 2024 12:39:18 +0800
Subject: [PATCH 05/18] make dimensions default

---
 metagpt/rag/schema.py | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/metagpt/rag/schema.py b/metagpt/rag/schema.py
index bedba164c..ccd727687 100644
--- a/metagpt/rag/schema.py
+++ b/metagpt/rag/schema.py
@@ -35,22 +35,23 @@ class IndexRetrieverConfig(BaseRetrieverConfig):
 class FAISSRetrieverConfig(IndexRetrieverConfig):
     """Config for FAISS-based retrievers."""
 
-    dimensions: int = Field(
-        default=config.embedding.dimensions, description="Dimensionality of the vectors for FAISS index construction."
-    )
+    dimensions: int = Field(default=0, description="Dimensionality of the vectors for FAISS index construction.")
 
     _embedding_type_to_dimensions: ClassVar[dict[EmbeddingType, int]] = {
         EmbeddingType.GEMINI: 768,
+        EmbeddingType.OLLAMA: 4096,
     }
 
     @model_validator(mode="after")
     def check_dimensions(self):
-        if self.dimensions is None:
+        if self.dimensions == 0:
+            self.dimensions = config.embedding.dimensions or self._embedding_type_to_dimensions.get(
+                config.embedding.api_type, 1536
+            )
             if config.embedding.api_type not in self._embedding_type_to_dimensions:
-                logger.info(
-                    f"You didn't set the dimensions in config when using {config.embedding.api_type}, default to 1536"
+                logger.warning(
+                    f"You didn't set dimensions in config when using {config.embedding.api_type}, default to 1536"
                 )
-            self.dimensions = self._embedding_type_to_dimensions.get(config.embedding.api_type, 1536)
 
         return self
 

From 01833aa997a33d1cf6989accda0d8a0ad318ce79 Mon Sep 17 00:00:00 2001
From: seehi <6580@pm.me>
Date: Tue, 7 May 2024 15:43:20 +0800
Subject: [PATCH 06/18] update comment in rag_pipeline example

---
 examples/rag_pipeline.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/rag_pipeline.py b/examples/rag_pipeline.py
index 75888e480..fbbc2cc39 100644
--- a/examples/rag_pipeline.py
+++ b/examples/rag_pipeline.py
@@ -42,7 +42,7 @@ class Player(BaseModel):
 class RAGExample:
     """Show how to use RAG."""
 
-    def __init__(self, engine: SimpleEngine = None, use_llm_ranker: bool = False):
+    def __init__(self, engine: SimpleEngine = None, use_llm_ranker: bool = True):
         self._engine = engine
         self._use_llm_ranker = use_llm_ranker
 

From d31d7507ef893e344af52e01ae07eace5b3962df Mon Sep 17 00:00:00 2001
From: usamimeri_renko <1710269958@qq.com>
Date: Tue, 7 May 2024 16:27:25 +0800
Subject: [PATCH 07/18] fix

---
 metagpt/rag/schema.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/metagpt/rag/schema.py b/metagpt/rag/schema.py
index ccd727687..618880a22 100644
--- a/metagpt/rag/schema.py
+++ b/metagpt/rag/schema.py
@@ -48,7 +48,7 @@ class FAISSRetrieverConfig(IndexRetrieverConfig):
             self.dimensions = config.embedding.dimensions or self._embedding_type_to_dimensions.get(
                 config.embedding.api_type, 1536
             )
-            if config.embedding.api_type not in self._embedding_type_to_dimensions:
+            if not config.embedding.dimensions and config.embedding.api_type not in self._embedding_type_to_dimensions:
                 logger.warning(
                     f"You didn't set dimensions in config when using {config.embedding.api_type}, default to 1536"
                 )

From 8695629fce6be6b72c53ec6c6f48d051bdcac6a0 Mon Sep 17 00:00:00 2001
From: huangpeilin <huangpeilin@qunjielong.com>
Date: Wed, 8 May 2024 19:22:10 +0800
Subject: [PATCH 08/18] =?UTF-8?q?fix:=20=E4=BF=AE=E5=A4=8Dusage=20?=
 =?UTF-8?q?=E5=80=BC=E4=B8=BAnone=20=E7=9A=84=E6=83=85=E5=86=B5?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 metagpt/provider/openai_api.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/metagpt/provider/openai_api.py b/metagpt/provider/openai_api.py
index 7957f775c..cc1a4fc8b 100644
--- a/metagpt/provider/openai_api.py
+++ b/metagpt/provider/openai_api.py
@@ -100,7 +100,7 @@ class OpenAILLM(BaseLLM):
             log_llm_stream(chunk_message)
             collected_messages.append(chunk_message)
             if finish_reason:
-                if hasattr(chunk, "usage"):
+                if hasattr(chunk, "usage") and chunk.usage:
                     # Some services have usage as an attribute of the chunk, such as Fireworks
                     usage = CompletionUsage(**chunk.usage)
                 elif hasattr(chunk.choices[0], "usage"):

From a473c74c902b4e9e1aa9ac21baaf61e2a92ba6d5 Mon Sep 17 00:00:00 2001
From: huangpeilin <huangpeilin@qunjielong.com>
Date: Thu, 9 May 2024 11:36:07 +0800
Subject: [PATCH 09/18] =?UTF-8?q?feat:=20=E4=BC=98=E5=8C=96=E5=88=A4?=
 =?UTF-8?q?=E6=96=AD=E9=80=BB=E8=BE=91?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 metagpt/provider/openai_api.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/metagpt/provider/openai_api.py b/metagpt/provider/openai_api.py
index cc1a4fc8b..68dc156c2 100644
--- a/metagpt/provider/openai_api.py
+++ b/metagpt/provider/openai_api.py
@@ -100,7 +100,7 @@ class OpenAILLM(BaseLLM):
             log_llm_stream(chunk_message)
             collected_messages.append(chunk_message)
             if finish_reason:
-                if hasattr(chunk, "usage") and chunk.usage:
+                if hasattr(chunk, "usage") and chunk.usage is not None:
                     # Some services have usage as an attribute of the chunk, such as Fireworks
                     usage = CompletionUsage(**chunk.usage)
                 elif hasattr(chunk.choices[0], "usage"):

From d4bd66898327b818eb80eb12a7bba25305304e79 Mon Sep 17 00:00:00 2001
From: Kunal Jain <kunal@joinwido.com>
Date: Sat, 11 May 2024 19:03:55 +0530
Subject: [PATCH 10/18] fix ask() got an unexpected keyword argument 'images'

---
 metagpt/provider/human_provider.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/metagpt/provider/human_provider.py b/metagpt/provider/human_provider.py
index 87dbd105f..a16a49c20 100644
--- a/metagpt/provider/human_provider.py
+++ b/metagpt/provider/human_provider.py
@@ -33,6 +33,7 @@ class HumanProvider(BaseLLM):
         format_msgs: Optional[list[dict[str, str]]] = None,
         generator: bool = False,
         timeout=USE_CONFIG_TIMEOUT,
+        **kwargs
     ) -> str:
         return self.ask(msg, timeout=self.get_timeout(timeout))
 

From 82b4be51bdd69000cc1ace7b7140c6c5f2e07ab9 Mon Sep 17 00:00:00 2001
From: usamimeri_renko <1710269958@qq.com>
Date: Mon, 13 May 2024 15:00:22 +0800
Subject: [PATCH 11/18] fix schema bug

---
 metagpt/provider/base_llm.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/metagpt/provider/base_llm.py b/metagpt/provider/base_llm.py
index 6387e3936..e554f7767 100644
--- a/metagpt/provider/base_llm.py
+++ b/metagpt/provider/base_llm.py
@@ -65,7 +65,7 @@ class BaseLLM(ABC):
             # image url or image base64
             url = image if image.startswith("http") else f"data:image/jpeg;base64,{image}"
             # it can with multiple-image inputs
-            content.append({"type": "image_url", "image_url": url})
+            content.append({"type": "image_url", "image_url": {"url": url}})
         return {"role": "user", "content": content}
 
     def _assistant_msg(self, msg: str) -> dict[str, str]:

From 133874cc0b2dcc30e759c61fadf036a0f3d0cec7 Mon Sep 17 00:00:00 2001
From: seehi <6580@pm.me>
Date: Tue, 14 May 2024 15:56:52 +0800
Subject: [PATCH 12/18] format

---
 metagpt/rag/benchmark/base.py   |  2 +-
 metagpt/rag/factories/ranker.py | 10 ++++++----
 2 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/metagpt/rag/benchmark/base.py b/metagpt/rag/benchmark/base.py
index c1fd297d9..b5d265b35 100644
--- a/metagpt/rag/benchmark/base.py
+++ b/metagpt/rag/benchmark/base.py
@@ -121,7 +121,7 @@ class RAGBenchmark:
                     return mrr_sum
 
         return mrr_sum
-        
+
     async def semantic_similarity(self, response: str, reference: str) -> float:
         result = await self.evaluator.aevaluate(
             response=response,
diff --git a/metagpt/rag/factories/ranker.py b/metagpt/rag/factories/ranker.py
index b75745a1f..7abda162a 100644
--- a/metagpt/rag/factories/ranker.py
+++ b/metagpt/rag/factories/ranker.py
@@ -8,11 +8,11 @@ from metagpt.rag.factories.base import ConfigBasedFactory
 from metagpt.rag.rankers.object_ranker import ObjectSortPostprocessor
 from metagpt.rag.schema import (
     BaseRankerConfig,
+    BGERerankConfig,
+    CohereRerankConfig,
     ColbertRerankConfig,
     LLMRankerConfig,
     ObjectRankerConfig,
-    CohereRerankConfig,
-    BGERerankConfig
 )
 
 
@@ -60,13 +60,15 @@ class RankerFactory(ConfigBasedFactory):
 
     def _create_bge_rerank(self, config: BGERerankConfig, **kwargs) -> LLMRerank:
         try:
-            from llama_index.postprocessor.flag_embedding_reranker import FlagEmbeddingReranker
+            from llama_index.postprocessor.flag_embedding_reranker import (
+                FlagEmbeddingReranker,
+            )
         except ImportError:
             raise ImportError(
                 "`llama-index-postprocessor-flag-embedding-reranker` package not found, please run `pip install llama-index-postprocessor-flag-embedding-reranker`"
             )
         return FlagEmbeddingReranker(**config.model_dump())
-        
+
     def _create_object_ranker(self, config: ObjectRankerConfig, **kwargs) -> LLMRerank:
         return ObjectSortPostprocessor(**config.model_dump())
 

From 1047abf1f40c6bc460247bba6254c42a6adaa04a Mon Sep 17 00:00:00 2001
From: usamimeri_renko <1710269958@qq.com>
Date: Wed, 15 May 2024 15:40:59 +0800
Subject: [PATCH 13/18] add counter

---
 metagpt/utils/token_counter.py | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/metagpt/utils/token_counter.py b/metagpt/utils/token_counter.py
index 724d49afc..0c7048331 100644
--- a/metagpt/utils/token_counter.py
+++ b/metagpt/utils/token_counter.py
@@ -37,6 +37,8 @@ TOKEN_COSTS = {
     "gpt-4-turbo": {"prompt": 0.01, "completion": 0.03},
     "gpt-4-vision-preview": {"prompt": 0.01, "completion": 0.03},  # TODO add extra image price calculator
     "gpt-4-1106-vision-preview": {"prompt": 0.01, "completion": 0.03},
+    "gpt-4o": {"prompt": 0.005, "completion": 0.015},
+    "gpt-4o-2024-05-13": {"prompt": 0.005, "completion": 0.015},
     "text-embedding-ada-002": {"prompt": 0.0004, "completion": 0.0},
     "glm-3-turbo": {"prompt": 0.0007, "completion": 0.0007},  # 128k version, prompt + completion tokens=0.005￥/k-tokens
     "glm-4": {"prompt": 0.014, "completion": 0.014},  # 128k version, prompt + completion tokens=0.1￥/k-tokens
@@ -56,11 +58,13 @@ TOKEN_COSTS = {
     "claude-3-opus-20240229": {"prompt": 0.015, "completion": 0.075},
     "yi-34b-chat-0205": {"prompt": 0.0003, "completion": 0.0003},
     "yi-34b-chat-200k": {"prompt": 0.0017, "completion": 0.0017},
+    "yi-large": {"prompt": 0.0028, "completion": 0.0028},
     "microsoft/wizardlm-2-8x22b": {"prompt": 0.00108, "completion": 0.00108},  # for openrouter, start
     "meta-llama/llama-3-70b-instruct": {"prompt": 0.008, "completion": 0.008},
     "llama3-70b-8192": {"prompt": 0.0059, "completion": 0.0079},
     "openai/gpt-3.5-turbo-0125": {"prompt": 0.0005, "completion": 0.0015},
     "openai/gpt-4-turbo-preview": {"prompt": 0.01, "completion": 0.03},
+    "deepseek-chat": {"prompt": 0.00014, "completion": 0.00028},
 }
 
 
@@ -224,6 +228,8 @@ def count_message_tokens(messages, model="gpt-3.5-turbo-0125"):
         "gpt-4-turbo",
         "gpt-4-vision-preview",
         "gpt-4-1106-vision-preview",
+        "gpt-4o-2024-05-13",
+        "gpt-4o",
     }:
         tokens_per_message = 3  # # every reply is primed with <|start|>assistant<|message|>
         tokens_per_name = 1

From 400e0af8144ebc6752adfe96334127d8395fef4c Mon Sep 17 00:00:00 2001
From: seehi <6580@pm.me>
Date: Wed, 15 May 2024 17:31:44 +0800
Subject: [PATCH 14/18] format

---
 metagpt/rag/benchmark/base.py   |  2 +-
 metagpt/rag/factories/ranker.py | 10 ++++++----
 2 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/metagpt/rag/benchmark/base.py b/metagpt/rag/benchmark/base.py
index c1fd297d9..b5d265b35 100644
--- a/metagpt/rag/benchmark/base.py
+++ b/metagpt/rag/benchmark/base.py
@@ -121,7 +121,7 @@ class RAGBenchmark:
                     return mrr_sum
 
         return mrr_sum
-        
+
     async def semantic_similarity(self, response: str, reference: str) -> float:
         result = await self.evaluator.aevaluate(
             response=response,
diff --git a/metagpt/rag/factories/ranker.py b/metagpt/rag/factories/ranker.py
index b75745a1f..7abda162a 100644
--- a/metagpt/rag/factories/ranker.py
+++ b/metagpt/rag/factories/ranker.py
@@ -8,11 +8,11 @@ from metagpt.rag.factories.base import ConfigBasedFactory
 from metagpt.rag.rankers.object_ranker import ObjectSortPostprocessor
 from metagpt.rag.schema import (
     BaseRankerConfig,
+    BGERerankConfig,
+    CohereRerankConfig,
     ColbertRerankConfig,
     LLMRankerConfig,
     ObjectRankerConfig,
-    CohereRerankConfig,
-    BGERerankConfig
 )
 
 
@@ -60,13 +60,15 @@ class RankerFactory(ConfigBasedFactory):
 
     def _create_bge_rerank(self, config: BGERerankConfig, **kwargs) -> LLMRerank:
         try:
-            from llama_index.postprocessor.flag_embedding_reranker import FlagEmbeddingReranker
+            from llama_index.postprocessor.flag_embedding_reranker import (
+                FlagEmbeddingReranker,
+            )
         except ImportError:
             raise ImportError(
                 "`llama-index-postprocessor-flag-embedding-reranker` package not found, please run `pip install llama-index-postprocessor-flag-embedding-reranker`"
             )
         return FlagEmbeddingReranker(**config.model_dump())
-        
+
     def _create_object_ranker(self, config: ObjectRankerConfig, **kwargs) -> LLMRerank:
         return ObjectSortPostprocessor(**config.model_dump())
 

From b71ea5f407c268a846eff5fabc0886c7e74a05b4 Mon Sep 17 00:00:00 2001
From: seehi <6580@pm.me>
Date: Wed, 15 May 2024 17:40:21 +0800
Subject: [PATCH 15/18] update comment

---
 examples/rag_pipeline.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/rag_pipeline.py b/examples/rag_pipeline.py
index fbbc2cc39..5b716ce03 100644
--- a/examples/rag_pipeline.py
+++ b/examples/rag_pipeline.py
@@ -243,7 +243,7 @@ async def main():
     """RAG pipeline.
 
     Note:
-    1. If `use_llm_ranker` is True, then will use LLM Reranker to get better result, but it is not always guaranteed that the output will be parseable for reranking,
+    1. If `use_llm_ranker` is True, then it will use LLM Reranker to get better result, but it is not always guaranteed that the output will be parseable for reranking,
        prefer `gpt-4-turbo`, otherwise might encounter `IndexError: list index out of range` or `ValueError: invalid literal for int() with base 10`.
     """
     e = RAGExample(use_llm_ranker=False)

From 3f54646691a8abf1ee28e26a667d76ba4ddc78d0 Mon Sep 17 00:00:00 2001
From: usamimeri_renko <1710269958@qq.com>
Date: Wed, 15 May 2024 17:44:20 +0800
Subject: [PATCH 16/18] add token max and deepseek-coder,gpt-4-turbo-2024-04-09

---
 metagpt/utils/token_counter.py | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/metagpt/utils/token_counter.py b/metagpt/utils/token_counter.py
index 0c7048331..8645e6380 100644
--- a/metagpt/utils/token_counter.py
+++ b/metagpt/utils/token_counter.py
@@ -35,6 +35,7 @@ TOKEN_COSTS = {
     "gpt-4-1106-preview": {"prompt": 0.01, "completion": 0.03},
     "gpt-4-0125-preview": {"prompt": 0.01, "completion": 0.03},
     "gpt-4-turbo": {"prompt": 0.01, "completion": 0.03},
+    "gpt-4-turbo-2024-04-09": {"prompt": 0.01, "completion": 0.03},
     "gpt-4-vision-preview": {"prompt": 0.01, "completion": 0.03},  # TODO add extra image price calculator
     "gpt-4-1106-vision-preview": {"prompt": 0.01, "completion": 0.03},
     "gpt-4o": {"prompt": 0.005, "completion": 0.015},
@@ -65,6 +66,7 @@ TOKEN_COSTS = {
     "openai/gpt-3.5-turbo-0125": {"prompt": 0.0005, "completion": 0.0015},
     "openai/gpt-4-turbo-preview": {"prompt": 0.01, "completion": 0.03},
     "deepseek-chat": {"prompt": 0.00014, "completion": 0.00028},
+    "deepseek-coder": {"prompt": 0.00014, "completion": 0.00028},
 }
 
 
@@ -159,6 +161,9 @@ FIREWORKS_GRADE_TOKEN_COSTS = {
 
 # https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo
 TOKEN_MAX = {
+    "gpt-4o-2024-05-13": 128000,
+    "gpt-4o": 128000,
+    "gpt-4-turbo-2024-04-09": 128000,
     "gpt-4-0125-preview": 128000,
     "gpt-4-turbo-preview": 128000,
     "gpt-4-1106-preview": 128000,
@@ -195,11 +200,14 @@ TOKEN_MAX = {
     "claude-3-opus-20240229": 200000,
     "yi-34b-chat-0205": 4000,
     "yi-34b-chat-200k": 200000,
+    "yi-large": 16385,
     "microsoft/wizardlm-2-8x22b": 65536,
     "meta-llama/llama-3-70b-instruct": 8192,
     "llama3-70b-8192": 8192,
     "openai/gpt-3.5-turbo-0125": 16385,
     "openai/gpt-4-turbo-preview": 128000,
+    "deepseek-chat": 32768,
+    "deepseek-coder": 16385,
 }
 
 

From a754c0bb8c8f6df9b97769f80cccab7df15f1bc1 Mon Sep 17 00:00:00 2001
From: voidking <voidking@qq.com>
Date: Thu, 16 May 2024 11:20:02 +0800
Subject: [PATCH 17/18] =?UTF-8?q?Pre-commit=E6=B7=BB=E5=8A=A0=E5=88=B0envi?=
 =?UTF-8?q?ronment?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 .github/workflows/pre-commit.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.github/workflows/pre-commit.yaml b/.github/workflows/pre-commit.yaml
index ed4bbb144..d350a87f1 100644
--- a/.github/workflows/pre-commit.yaml
+++ b/.github/workflows/pre-commit.yaml
@@ -11,6 +11,7 @@ on:
 jobs:
   pre-commit-check:
     runs-on: ubuntu-latest
+    environment: pre-commit
     steps:
     - name: Checkout Source Code
       uses: actions/checkout@v2

From b1c1dd24f3f7509b9f3cbb8c74232b3ce0ae444f Mon Sep 17 00:00:00 2001
From: usamimeri_renko <93753250+usamimeri@users.noreply.github.com>
Date: Thu, 16 May 2024 15:20:36 +0800
Subject: [PATCH 18/18] Update requirements.txt

update tiktoken to 0.7.0 to support gpt-4o
---
 requirements.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/requirements.txt b/requirements.txt
index 6c219a9dc..2b8e1e13f 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -26,7 +26,7 @@ PyYAML==6.0.1
 # sentence_transformers==2.2.2
 setuptools==65.6.3
 tenacity==8.2.3
-tiktoken==0.6.0
+tiktoken==0.7.0
 tqdm==4.66.2
 #unstructured[local-inference]
 # selenium>4