letta-nightly 0.6.3.dev20241211050151__tar.gz → 0.6.3.dev20241212015858__tar.gz

{letta_nightly-0.6.3.dev20241211050151 → letta_nightly-0.6.3.dev20241212015858}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: letta-nightly
-Version: 0.6.3.dev20241211050151
+Version: 0.6.3.dev20241212015858
 Summary: Create LLM agents with long-term memory and custom tools
 License: Apache License
 Author: Letta Team
@@ -82,9 +82,9 @@ Description-Content-Type: text/markdown
 
 <p align="center">
   <picture>
-    <source media="(prefers-color-scheme: dark)" srcset="assets/Letta-logo-RGB_GreyonTransparent_cropped_small.png">
-    <source media="(prefers-color-scheme: light)" srcset="assets/Letta-logo-RGB_OffBlackonTransparent_cropped_small.png">
-    <img alt="Letta logo" src="assets/Letta-logo-RGB_GreyonOffBlack_cropped_small.png" width="500">
+    <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/Letta-logo-RGB_GreyonTransparent_cropped_small.png">
+    <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/Letta-logo-RGB_OffBlackonTransparent_cropped_small.png">
+    <img alt="Letta logo" src="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/Letta-logo-RGB_GreyonOffBlack_cropped_small.png" width="500">
   </picture>
 </p>
 
@@ -95,9 +95,9 @@ Description-Content-Type: text/markdown
 
 <p align="center">
   <picture>
-    <source media="(prefers-color-scheme: dark)" srcset="assets/example_ade_screenshot.png">
-    <source media="(prefers-color-scheme: light)" srcset="assets/example_ade_screenshot_light.png">
-    <img alt="Letta logo" src="assets/example_ade_screenshot.png" width="800">
+    <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot.png">
+    <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot_light.png">
+    <img alt="Letta logo" src="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot.png" width="800">
   </picture>
 </p>
 
@@ -133,7 +133,7 @@ Description-Content-Type: text/markdown
 
 ## ⚡ Quickstart
 
-_The recommended way to use Letta is to run use Docker. To install Docker, see [Docker's installation guide](https://docs.docker.com/get-docker/). For issues with installing Docker, see [Docker's troubleshooting guide](https://docs.docker.com/desktop/troubleshoot-and-support/troubleshoot/). You can also install Letta using `pip` (see guide [below](#-quickstart-pip))._
+_The recommended way to use Letta is to run use Docker. To install Docker, see [Docker's installation guide](https://docs.docker.com/get-docker/). For issues with installing Docker, see [Docker's troubleshooting guide](https://docs.docker.com/desktop/troubleshoot-and-support/troubleshoot/). You can also install Letta using `pip` (see instructions [below](#-quickstart-pip))._
 
 ### 🌖 Run the Letta server
 
@@ -171,9 +171,9 @@ Once the Letta server is running, you can access it via port `8283` (e.g. sendin
 
 <p align="center">
   <picture>
-    <source media="(prefers-color-scheme: dark)" srcset="assets/example_ade_screenshot.png">
-    <source media="(prefers-color-scheme: light)" srcset="assets/example_ade_screenshot_light.png">
-    <img alt="Letta logo" src="assets/example_ade_screenshot.png" width="800">
+    <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot.png">
+    <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot_light.png">
+    <img alt="ADE screenshot" src="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot.png" width="800">
   </picture>
 </p>
 
@@ -186,9 +186,9 @@ To connect the ADE with your local Letta server, simply:
 
 <p align="center">
   <picture>
-    <source media="(prefers-color-scheme: dark)" srcset="assets/example_ade_screenshot_agents.png">
-    <source media="(prefers-color-scheme: light)" srcset="assets/example_ade_screenshot_agents_light.png">
-    <img alt="Letta logo" src="assets/example_ade_screenshot_agents.png" width="800">
+    <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot_agents.png">
+    <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot_agents_light.png">
+    <img alt="Letta logo" src="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot_agents.png" width="800">
   </picture>
 </p>
 
@@ -217,6 +217,18 @@ If your Letta server isn't running on `localhost` (for example, you deployed it
 
 No, you can install Letta using `pip` (via `pip install -U letta`), as well as from source (via `poetry install`). See instructions below.
 
+> _"What's the difference between installing with `pip` vs `Docker`?"_
+
+Letta gives your agents persistence (they live indefinitely) by storing all your agent data in a database. Letta is designed to be used with a [PostgreSQL](https://en.wikipedia.org/wiki/PostgreSQL) (the world's most popular database), however, it is not possible to install PostgreSQL via `pip`, so the `pip` install of Letta defaults to using [SQLite](https://www.sqlite.org/). If you have a PostgreSQL instance running on your own computer, you can still connect Letta (installed via `pip`) to PostgreSQL by setting the environment variable `LETTA_PG_URI`.
+
+**Database migrations are not officially supported for Letta when using SQLite**, so you would like to ensure that if you're able to upgrade to the latest Letta version and migrate your Letta agents data, make sure that you're using PostgreSQL as your Letta database backend. Full compatability table below:
+
+| Installation method | Start server command | Database backend | Data migrations supported? |
+|---|---|---|---|
+| `pip install letta` | `letta server` | SQLite | ❌ |
+| `pip install letta` | `export LETTA_PG_URI=...` + `letta server` | PostgreSQL | ✅ |
+| *[Install Docker](https://www.docker.com/get-started/)*  |`docker run ...` ([full command](#-run-the-letta-server)) | PostgreSQL | ✅ |
+
 > _"How do I use the ADE locally?"_
 
 To connect the ADE to your local Letta server, simply run your Letta server (make sure you can access `localhost:8283`) and go to [https://app.letta.com](https://app.letta.com). If you would like to use the old version of the ADE (that runs on `localhost`), downgrade to Letta version `<=0.5.0`.

{letta_nightly-0.6.3.dev20241211050151 → letta_nightly-0.6.3.dev20241212015858}/README.md

@@ -1,8 +1,8 @@
 <p align="center">
   <picture>
-    <source media="(prefers-color-scheme: dark)" srcset="assets/Letta-logo-RGB_GreyonTransparent_cropped_small.png">
-    <source media="(prefers-color-scheme: light)" srcset="assets/Letta-logo-RGB_OffBlackonTransparent_cropped_small.png">
-    <img alt="Letta logo" src="assets/Letta-logo-RGB_GreyonOffBlack_cropped_small.png" width="500">
+    <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/Letta-logo-RGB_GreyonTransparent_cropped_small.png">
+    <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/Letta-logo-RGB_OffBlackonTransparent_cropped_small.png">
+    <img alt="Letta logo" src="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/Letta-logo-RGB_GreyonOffBlack_cropped_small.png" width="500">
   </picture>
 </p>
 
@@ -13,9 +13,9 @@
 
 <p align="center">
   <picture>
-    <source media="(prefers-color-scheme: dark)" srcset="assets/example_ade_screenshot.png">
-    <source media="(prefers-color-scheme: light)" srcset="assets/example_ade_screenshot_light.png">
-    <img alt="Letta logo" src="assets/example_ade_screenshot.png" width="800">
+    <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot.png">
+    <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot_light.png">
+    <img alt="Letta logo" src="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot.png" width="800">
   </picture>
 </p>
 
@@ -51,7 +51,7 @@
 
 ## ⚡ Quickstart
 
-_The recommended way to use Letta is to run use Docker. To install Docker, see [Docker's installation guide](https://docs.docker.com/get-docker/). For issues with installing Docker, see [Docker's troubleshooting guide](https://docs.docker.com/desktop/troubleshoot-and-support/troubleshoot/). You can also install Letta using `pip` (see guide [below](#-quickstart-pip))._
+_The recommended way to use Letta is to run use Docker. To install Docker, see [Docker's installation guide](https://docs.docker.com/get-docker/). For issues with installing Docker, see [Docker's troubleshooting guide](https://docs.docker.com/desktop/troubleshoot-and-support/troubleshoot/). You can also install Letta using `pip` (see instructions [below](#-quickstart-pip))._
 
 ### 🌖 Run the Letta server
 
@@ -89,9 +89,9 @@ Once the Letta server is running, you can access it via port `8283` (e.g. sendin
 
 <p align="center">
   <picture>
-    <source media="(prefers-color-scheme: dark)" srcset="assets/example_ade_screenshot.png">
-    <source media="(prefers-color-scheme: light)" srcset="assets/example_ade_screenshot_light.png">
-    <img alt="Letta logo" src="assets/example_ade_screenshot.png" width="800">
+    <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot.png">
+    <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot_light.png">
+    <img alt="ADE screenshot" src="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot.png" width="800">
   </picture>
 </p>
 
@@ -104,9 +104,9 @@ To connect the ADE with your local Letta server, simply:
 
 <p align="center">
   <picture>
-    <source media="(prefers-color-scheme: dark)" srcset="assets/example_ade_screenshot_agents.png">
-    <source media="(prefers-color-scheme: light)" srcset="assets/example_ade_screenshot_agents_light.png">
-    <img alt="Letta logo" src="assets/example_ade_screenshot_agents.png" width="800">
+    <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot_agents.png">
+    <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot_agents_light.png">
+    <img alt="Letta logo" src="https://raw.githubusercontent.com/letta-ai/letta/refs/heads/main/assets/example_ade_screenshot_agents.png" width="800">
   </picture>
 </p>
 
@@ -135,6 +135,18 @@ If your Letta server isn't running on `localhost` (for example, you deployed it
 
 No, you can install Letta using `pip` (via `pip install -U letta`), as well as from source (via `poetry install`). See instructions below.
 
+> _"What's the difference between installing with `pip` vs `Docker`?"_
+
+Letta gives your agents persistence (they live indefinitely) by storing all your agent data in a database. Letta is designed to be used with a [PostgreSQL](https://en.wikipedia.org/wiki/PostgreSQL) (the world's most popular database), however, it is not possible to install PostgreSQL via `pip`, so the `pip` install of Letta defaults to using [SQLite](https://www.sqlite.org/). If you have a PostgreSQL instance running on your own computer, you can still connect Letta (installed via `pip`) to PostgreSQL by setting the environment variable `LETTA_PG_URI`.
+
+**Database migrations are not officially supported for Letta when using SQLite**, so you would like to ensure that if you're able to upgrade to the latest Letta version and migrate your Letta agents data, make sure that you're using PostgreSQL as your Letta database backend. Full compatability table below:
+
+| Installation method | Start server command | Database backend | Data migrations supported? |
+|---|---|---|---|
+| `pip install letta` | `letta server` | SQLite | ❌ |
+| `pip install letta` | `export LETTA_PG_URI=...` + `letta server` | PostgreSQL | ✅ |
+| *[Install Docker](https://www.docker.com/get-started/)*  |`docker run ...` ([full command](#-run-the-letta-server)) | PostgreSQL | ✅ |
+
 > _"How do I use the ADE locally?"_
 
 To connect the ADE to your local Letta server, simply run your Letta server (make sure you can access `localhost:8283`) and go to [https://app.letta.com](https://app.letta.com). If you would like to use the old version of the ADE (that runs on `localhost`), downgrade to Letta version `<=0.5.0`.

{letta_nightly-0.6.3.dev20241211050151 → letta_nightly-0.6.3.dev20241212015858}/letta/agent.py

@@ -8,7 +8,6 @@ from typing import List, Literal, Optional, Tuple, Union
 
 from tqdm import tqdm
 
-from letta.agent_store.storage import StorageConnector
 from letta.constants import (
     BASE_TOOLS,
     CLI_WARNING_PREFIX,

{letta_nightly-0.6.3.dev20241211050151 → letta_nightly-0.6.3.dev20241212015858}/letta/data_sources/connectors.py

@@ -2,7 +2,6 @@ from typing import Dict, Iterator, List, Tuple, Optional
 
 import typer
 
-from letta.agent_store.storage import StorageConnector
 from letta.data_sources.connectors_helper import (
     assert_all_files_exist_locally,
     extract_metadata_from_files,
@@ -12,6 +11,7 @@ from letta.embeddings import embedding_model
 from letta.schemas.file import FileMetadata
 from letta.schemas.passage import Passage
 from letta.schemas.source import Source
+from letta.services.passage_manager import PassageManager
 from letta.services.source_manager import SourceManager
 from letta.utils import create_uuid_from_string
 
@@ -42,7 +42,7 @@ class DataConnector:
         """
 
 
-def load_data(connector: DataConnector, source: Source, passage_store: StorageConnector, source_manager: SourceManager, actor: "User", agent_id: Optional[str] = None):
+def load_data(connector: DataConnector, source: Source, passage_manager: PassageManager, source_manager: SourceManager, actor: "User", agent_id: Optional[str] = None):
     """Load data from a connector (generates file and passages) into a specified source_id, associated with a user_id."""
     embedding_config = source.embedding_config
 
@@ -103,14 +103,14 @@ def load_data(connector: DataConnector, source: Source, passage_store: StorageCo
             embedding_to_document_name[hashable_embedding] = file_name
             if len(passages) >= 100:
                 # insert passages into passage store
-                passage_store.insert_many(passages)
+                passage_manager.create_many_passages(passages, actor)
 
                 passage_count += len(passages)
                 passages = []
 
     if len(passages) > 0:
         # insert passages into passage store
-        passage_store.insert_many(passages)
+        passage_manager.create_many_passages(passages, actor)
         passage_count += len(passages)
 
     return passage_count, file_count
@@ -170,74 +170,3 @@ class DirectoryConnector(DataConnector):
         nodes = parser.get_nodes_from_documents(documents)
         for node in nodes:
             yield node.text, None
-
-
-"""
-The below isn't used anywhere, it isn't tested, and pretty much should be deleted.
-- Matt
-"""
-# class WebConnector(DirectoryConnector):
-#     def __init__(self, urls: List[str] = None, html_to_text: bool = True):
-#         self.urls = urls
-#         self.html_to_text = html_to_text
-#
-#     def generate_files(self) -> Iterator[Tuple[str, Dict]]:  # -> Iterator[Document]:
-#         from llama_index.readers.web import SimpleWebPageReader
-#
-#         files = SimpleWebPageReader(html_to_text=self.html_to_text).load_data(self.urls)
-#         for document in files:
-#             yield document.text, {"url": document.id_}
-#
-#
-# class VectorDBConnector(DataConnector):
-#     # NOTE: this class has not been properly tested, so is unlikely to work
-#     # TODO: allow loading multiple tables (1:1 mapping between FileMetadata and Table)
-#
-#     def __init__(
-#         self,
-#         name: str,
-#         uri: str,
-#         table_name: str,
-#         text_column: str,
-#         embedding_column: str,
-#         embedding_dim: int,
-#     ):
-#         self.name = name
-#         self.uri = uri
-#         self.table_name = table_name
-#         self.text_column = text_column
-#         self.embedding_column = embedding_column
-#         self.embedding_dim = embedding_dim
-#
-#         # connect to db table
-#         from sqlalchemy import create_engine
-#
-#         self.engine = create_engine(uri)
-#
-#     def generate_files(self) -> Iterator[Tuple[str, Dict]]:  # -> Iterator[Document]:
-#         yield self.table_name, None
-#
-#     def generate_passages(self, file_text: str, file: FileMetadata, chunk_size: int = 1024) -> Iterator[Tuple[str, Dict]]:  # -> Iterator[Passage]:
-#         from pgvector.sqlalchemy import Vector
-#         from sqlalchemy import Inspector, MetaData, Table, select
-#
-#         metadata = MetaData()
-#         # Create an inspector to inspect the database
-#         inspector = Inspector.from_engine(self.engine)
-#         table_names = inspector.get_table_names()
-#         assert self.table_name in table_names, f"Table {self.table_name} not found in database: tables that exist {table_names}."
-#
-#         table = Table(self.table_name, metadata, autoload_with=self.engine)
-#
-#         # Prepare a select statement
-#         select_statement = select(table.c[self.text_column], table.c[self.embedding_column].cast(Vector(self.embedding_dim)))
-#
-#         # Execute the query and fetch the results
-#         # TODO: paginate results
-#         with self.engine.connect() as connection:
-#             result = connection.execute(select_statement).fetchall()
-#
-#         for text, embedding in result:
-#             # assume that embeddings are the same model as in config
-#             # TODO: don't re-compute embedding
-#             yield text, {"embedding": embedding}

letta_nightly-0.6.3.dev20241212015858/letta/memory.py

@@ -0,0 +1,89 @@
+import datetime
+from abc import ABC, abstractmethod
+from typing import Callable, Dict, List, Tuple, Union
+
+from letta.constants import MESSAGE_SUMMARY_REQUEST_ACK, MESSAGE_SUMMARY_WARNING_FRAC
+from letta.embeddings import embedding_model, parse_and_chunk_text, query_embedding
+from letta.llm_api.llm_api_tools import create
+from letta.prompts.gpt_summarize import SYSTEM as SUMMARY_PROMPT_SYSTEM
+from letta.schemas.agent import AgentState
+from letta.schemas.enums import MessageRole
+from letta.schemas.memory import Memory
+from letta.schemas.message import Message
+from letta.schemas.passage import Passage
+from letta.utils import (
+    count_tokens,
+    extract_date_from_timestamp,
+    get_local_time,
+    printd,
+    validate_date_format,
+)
+
+
+def get_memory_functions(cls: Memory) -> Dict[str, Callable]:
+    """Get memory functions for a memory class"""
+    functions = {}
+
+    # collect base memory functions (should not be included)
+    base_functions = []
+    for func_name in dir(Memory):
+        funct = getattr(Memory, func_name)
+        if callable(funct):
+            base_functions.append(func_name)
+
+    for func_name in dir(cls):
+        if func_name.startswith("_") or func_name in ["load", "to_dict"]:  # skip base functions
+            continue
+        if func_name in base_functions:  # dont use BaseMemory functions
+            continue
+        func = getattr(cls, func_name)
+        if not callable(func):  # not a function
+            continue
+        functions[func_name] = func
+    return functions
+
+
+def _format_summary_history(message_history: List[Message]):
+    # TODO use existing prompt formatters for this (eg ChatML)
+    return "\n".join([f"{m.role}: {m.text}" for m in message_history])
+
+
+def summarize_messages(
+    agent_state: AgentState,
+    message_sequence_to_summarize: List[Message],
+):
+    """Summarize a message sequence using GPT"""
+    # we need the context_window
+    context_window = agent_state.llm_config.context_window
+
+    summary_prompt = SUMMARY_PROMPT_SYSTEM
+    summary_input = _format_summary_history(message_sequence_to_summarize)
+    summary_input_tkns = count_tokens(summary_input)
+    if summary_input_tkns > MESSAGE_SUMMARY_WARNING_FRAC * context_window:
+        trunc_ratio = (MESSAGE_SUMMARY_WARNING_FRAC * context_window / summary_input_tkns) * 0.8  # For good measure...
+        cutoff = int(len(message_sequence_to_summarize) * trunc_ratio)
+        summary_input = str(
+            [summarize_messages(agent_state, message_sequence_to_summarize=message_sequence_to_summarize[:cutoff])]
+            + message_sequence_to_summarize[cutoff:]
+        )
+
+    agent_state.user_id
+    dummy_agent_id = agent_state.id
+    message_sequence = []
+    message_sequence.append(Message(agent_id=dummy_agent_id, role=MessageRole.system, text=summary_prompt))
+    message_sequence.append(Message(agent_id=dummy_agent_id, role=MessageRole.assistant, text=MESSAGE_SUMMARY_REQUEST_ACK))
+    message_sequence.append(Message(agent_id=dummy_agent_id, role=MessageRole.user, text=summary_input))
+
+    # TODO: We need to eventually have a separate LLM config for the summarizer LLM
+    llm_config_no_inner_thoughts = agent_state.llm_config.model_copy(deep=True)
+    llm_config_no_inner_thoughts.put_inner_thoughts_in_kwargs = False
+    response = create(
+        llm_config=llm_config_no_inner_thoughts,
+        user_id=agent_state.user_id,
+        messages=message_sequence,
+        stream=False,
+    )
+
+    printd(f"summarize_messages gpt reply: {response.choices[0]}")
+    reply = response.choices[0].message.content
+    return reply

{letta_nightly-0.6.3.dev20241211050151 → letta_nightly-0.6.3.dev20241212015858}/letta/schemas/message.py

@@ -4,7 +4,7 @@ import warnings
 from datetime import datetime, timezone
 from typing import List, Literal, Optional
 
-from pydantic import Field, field_validator
+from pydantic import BaseModel, Field, field_validator
 
 from letta.constants import (
     DEFAULT_MESSAGE_TOOL,
@@ -54,7 +54,7 @@ class BaseMessage(OrmMetadataBase):
     __id_prefix__ = "message"
 
 
-class MessageCreate(BaseMessage):
+class MessageCreate(BaseModel):
     """Request to create a message"""
 
     # In the simplified format, only allow simple roles
@@ -66,7 +66,7 @@ class MessageCreate(BaseMessage):
     name: Optional[str] = Field(None, description="The name of the participant.")
 
 
-class MessageUpdate(BaseMessage):
+class MessageUpdate(BaseModel):
     """Request to update a message"""
 
     role: Optional[MessageRole] = Field(None, description="The role of the participant.")

{letta_nightly-0.6.3.dev20241211050151 → letta_nightly-0.6.3.dev20241212015858}/letta/server/server.py

@@ -16,7 +16,6 @@ import letta.constants as constants
 import letta.server.utils as server_utils
 import letta.system as system
 from letta.agent import Agent, save_agent
-from letta.agent_store.storage import StorageConnector, TableType
 from letta.chat_only_agent import ChatOnlyAgent
 from letta.credentials import LettaCredentials
 from letta.data_sources.connectors import DataConnector, load_data
@@ -558,7 +557,6 @@ class SyncServer(Server):
                 raise ValueError(command)
 
             # attach data to agent from source
-            source_connector = StorageConnector.get_storage_connector(TableType.PASSAGES, self.config, user_id=user_id)
             letta_agent.attach_source(
                 user=self.user_manager.get_user_by_id(user_id=user_id),
                 source_id=data_source,
@@ -1571,8 +1569,7 @@ class SyncServer(Server):
         self.source_manager.delete_source(source_id=source_id, actor=actor)
 
         # delete data from passage store
-        passage_store = StorageConnector.get_storage_connector(TableType.PASSAGES, self.config, user_id=actor.id)
-        passage_store.delete({"source_id": source_id})
+        self.passage_manager.delete_passages(actor=actor, limit=None, source_id=source_id)
 
         # TODO: delete data from agent passage stores (?)
 
@@ -1614,11 +1611,10 @@ class SyncServer(Server):
         if source is None:
             raise ValueError(f"Data source {source_name} does not exist for user {user_id}")
 
-        # get the data connectors
-        passage_store = StorageConnector.get_storage_connector(TableType.PASSAGES, self.config, user_id=user_id)
-
         # load data into the document store
-        passage_count, document_count = load_data(connector, source, passage_store, self.source_manager, actor=user, agent_id=agent_id)
+        passage_count, document_count = load_data(
+            connector, source, self.passage_manager, self.source_manager, actor=user, agent_id=agent_id
+        )
         return passage_count, document_count
 
     def attach_source_to_agent(
@@ -1694,8 +1690,7 @@ class SyncServer(Server):
         for source in sources:
 
             # count number of passages
-            passage_conn = StorageConnector.get_storage_connector(TableType.PASSAGES, self.config, user_id=actor.id)
-            num_passages = passage_conn.size({"source_id": source.id})
+            num_passages = self.passage_manager.size(actor=actor, source_id=source.id)
 
             # TODO: add when files table implemented
             ## count number of files
@@ -1806,14 +1801,20 @@ class SyncServer(Server):
 
         llm_models = []
         for provider in self._enabled_providers:
-            llm_models.extend(provider.list_llm_models())
+            try:
+                llm_models.extend(provider.list_llm_models())
+            except Exception as e:
+                warnings.warn(f"An error occurred while listing LLM models for provider {provider}: {e}")
         return llm_models
 
     def list_embedding_models(self) -> List[EmbeddingConfig]:
         """List available embedding models"""
         embedding_models = []
         for provider in self._enabled_providers:
-            embedding_models.extend(provider.list_embedding_models())
+            try:
+                embedding_models.extend(provider.list_embedding_models())
+            except Exception as e:
+                warnings.warn(f"An error occurred while listing embedding models for provider {provider}: {e}")
         return embedding_models
 
     def add_llm_model(self, request: LLMConfig) -> LLMConfig:

{letta_nightly-0.6.3.dev20241211050151 → letta_nightly-0.6.3.dev20241212015858}/letta/settings.py

@@ -17,7 +17,7 @@ class ToolSettings(BaseSettings):
 
 class ModelSettings(BaseSettings):
 
-    model_config = SettingsConfigDict(env_file='.env')
+    model_config = SettingsConfigDict(env_file='.env', extra='ignore')
 
     # env_prefix='my_prefix_'
 
@@ -64,7 +64,7 @@ cors_origins = ["http://letta.localhost", "http://localhost:8283", "http://local
 
 
 class Settings(BaseSettings):
-    model_config = SettingsConfigDict(env_prefix="letta_")
+    model_config = SettingsConfigDict(env_prefix="letta_", extra='ignore')
 
     letta_dir: Optional[Path] = Field(Path.home() / ".letta", env="LETTA_DIR")
     debug: Optional[bool] = False
@@ -103,7 +103,7 @@ class Settings(BaseSettings):
 
 
 class TestSettings(Settings):
-    model_config = SettingsConfigDict(env_prefix="letta_test_")
+    model_config = SettingsConfigDict(env_prefix="letta_test_", extra='ignore')
 
     letta_dir: Optional[Path] = Field(Path.home() / ".letta/test", env="LETTA_TEST_DIR")
 

{letta_nightly-0.6.3.dev20241211050151 → letta_nightly-0.6.3.dev20241212015858}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "letta-nightly"
-version = "0.6.3.dev20241211050151"
+version = "0.6.3.dev20241212015858"
 packages = [
     {include = "letta"}
 ]