letta-code cli failed to call local letta server with Ollama only

[ansidev]

Reproduction steps

1. OS: macOS
2. Setup letta-server

docker run \
  -d \
  --name letta-server \
  -v ~/.letta/.persist/pgdata:/var/lib/postgresql/data \
  -p 8283:8283 \
  -e OLLAMA_BASE_URL="http://host.docker.internal:{{.OLLAMA_PORT}}/v1" \
  letta/letta:latest

3. Run letta CLI

LETTA_BASE_URL="http://localhost:8283" letta

4. Verified ollama models are updated in letta CLI using /model.
5. Send any message
6. Server response:

⚠ {
    "error": {
      "error": {
        "message_type": "error_message",
        "run_id": "run-f2c55435-eaa9-4b5d-b2f9-30f7a9ca052b",
        "error_type": "llm_error",
        "message": "An error occurred with the LLM request.",
        "detail": "Unhandled LLM error: Invalid port: 'http:'",
        "seq_id": null
      },
      "run_id": "run-f2c55435-eaa9-4b5d-b2f9-30f7a9ca052b"
    }
  }

⚠ Downstream provider issues? Use /model to switch to another provider

7. Server logs:

2026-02-23 07:29:26.879 UTC [40] LOG:  checkpoint starting: time
2026-02-23 07:29:30.232 UTC [40] LOG:  checkpoint complete: wrote 35 buffers (0.2%); 0 WAL file(s) added, 0 removed, 0 recycled; write=3.338 s, sync=0.008 s, total=3.354 s; sync files=25, longest=0.002 s, average=0.001 s; distance=99 kB, estimate=180 kB
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - WARNING - Context token estimate is not set
Letta.letta.orm.sqlalchemy_base - WARNING - SECURITY: Listing org-scoped model Step without actor. This bypasses organization filtering.
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - WARNING - Error during step processing: Unhandled LLM error: Invalid port: 'http:'
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - INFO - Running final update. Step Progression: StepProgression.START
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - WARNING - Error during agent stream: Unhandled LLM error: Invalid port: 'http:'
Traceback (most recent call last):
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urlparse.py", line 409, in normalize_port
    port_as_int = int(port)
                  ^^^^^^^^^
ValueError: invalid literal for int() with base 10: 'http:'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/app/letta/adapters/simple_llm_stream_adapter.py", line 139, in invoke_llm
    stream = await self.llm_client.stream_async(request_data, self.llm_config)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/letta/otel/tracing.py", line 393, in async_wrapper
    return await func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/letta/llm_api/openai_client.py", line 793, in stream_async
    client = AsyncOpenAI(**kwargs)
             ^^^^^^^^^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/openai/_client.py", line 517, in __init__
    super().__init__(
  File "/app/.venv/lib/python3.11/site-packages/openai/_base_client.py", line 1399, in __init__
    super().__init__(
  File "/app/.venv/lib/python3.11/site-packages/openai/_base_client.py", line 383, in __init__
    self._base_url = self._enforce_trailing_slash(URL(base_url))
                                                  ^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urls.py", line 117, in __init__
    self._uri_reference = urlparse(url, **kwargs)
                          ^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urlparse.py", line 321, in urlparse
    parsed_port: int | None = normalize_port(port, scheme)
                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urlparse.py", line 411, in normalize_port
    raise InvalidURL(f"Invalid port: {port!r}")
httpx.InvalidURL: Invalid port: 'http:'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/app/letta/agents/letta_agent_v3.py", line 377, in stream
    async for chunk in response:
  File "/app/letta/agents/letta_agent_v3.py", line 960, in _step
    raise e
  File "/app/letta/agents/letta_agent_v3.py", line 786, in _step
    raise e
  File "/app/letta/agents/letta_agent_v3.py", line 774, in _step
    async for chunk in invocation:
  File "/app/letta/adapters/simple_llm_stream_adapter.py", line 141, in invoke_llm
    raise self.llm_client.handle_llm_error(e)
letta.errors.LLMError: Unhandled LLM error: Invalid port: 'http:'
Letta.letta.services.streaming_service - ERROR - Run run-3830b891-7439-407b-a5d6-04b31cc6f114 stopped with LLM error: Unhandled LLM error: Invalid port: 'http:', error_data: {'message_type': 'error_message', 'run_id': 'run-3830b891-7439-407b-a5d6-04b31cc6f114', 'error_type': 'llm_error', 'message': 'An error occurred with the LLM request.', 'detail': "Unhandled LLM error: Invalid port: 'http:'", 'seq_id': None}
Letta.letta.services.run_manager - WARNING - Run run-3830b891-7439-407b-a5d6-04b31cc6f114 completed without a completed_at timestamp
Letta.letta.services.run_manager - ERROR - Run run-3830b891-7439-407b-a5d6-04b31cc6f114 is already in a terminal state failed with stop reason llm_api_error, but is being updated with data {'status': <RunStatus.failed: 'failed'>, 'completed_at': None, 'stop_reason': <StopReasonType.llm_api_error: 'llm_api_error'>, 'metadata': {'error': {'message_type': 'error_message', 'run_id': 'run-3830b891-7439-407b-a5d6-04b31cc6f114', 'error_type': 'llm_error', 'message': 'An error occurred with the LLM request.', 'detail': "Unhandled LLM error: Invalid port: 'http:'", 'seq_id': None}}, 'total_duration_ns': None}
Letta.letta.orm.sqlalchemy_base - WARNING - SECURITY: Listing org-scoped model Step without actor. This bypasses organization filtering.
Letta.letta.orm.sqlalchemy_base - WARNING - SECURITY: Listing org-scoped model Step without actor. This bypasses organization filtering.
Letta.letta.server.rest_api.redis_stream_manager - WARNING - [Stream Finalizer] Appending forced [DONE] for run=run-3830b891-7439-407b-a5d6-04b31cc6f114 (saw_error=True, saw_done=False, final_stop_reason=llm_api_error)
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - WARNING - Context token estimate is not set
Letta.letta.orm.sqlalchemy_base - WARNING - SECURITY: Listing org-scoped model Step without actor. This bypasses organization filtering.
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - WARNING - Error during step processing: Unhandled LLM error: Invalid port: 'http:'
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - INFO - Running final update. Step Progression: StepProgression.START
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - WARNING - Error during agent stream: Unhandled LLM error: Invalid port: 'http:'
Traceback (most recent call last):
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urlparse.py", line 409, in normalize_port
    port_as_int = int(port)
                  ^^^^^^^^^
ValueError: invalid literal for int() with base 10: 'http:'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/app/letta/adapters/simple_llm_stream_adapter.py", line 139, in invoke_llm
    stream = await self.llm_client.stream_async(request_data, self.llm_config)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/letta/otel/tracing.py", line 393, in async_wrapper
    return await func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/letta/llm_api/openai_client.py", line 793, in stream_async
    client = AsyncOpenAI(**kwargs)
             ^^^^^^^^^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/openai/_client.py", line 517, in __init__
    super().__init__(
  File "/app/.venv/lib/python3.11/site-packages/openai/_base_client.py", line 1399, in __init__
    super().__init__(
  File "/app/.venv/lib/python3.11/site-packages/openai/_base_client.py", line 383, in __init__
    self._base_url = self._enforce_trailing_slash(URL(base_url))
                                                  ^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urls.py", line 117, in __init__
    self._uri_reference = urlparse(url, **kwargs)
                          ^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urlparse.py", line 321, in urlparse
    parsed_port: int | None = normalize_port(port, scheme)
                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urlparse.py", line 411, in normalize_port
    raise InvalidURL(f"Invalid port: {port!r}")
httpx.InvalidURL: Invalid port: 'http:'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/app/letta/agents/letta_agent_v3.py", line 377, in stream
    async for chunk in response:
  File "/app/letta/agents/letta_agent_v3.py", line 960, in _step
    raise e
  File "/app/letta/agents/letta_agent_v3.py", line 786, in _step
    raise e
  File "/app/letta/agents/letta_agent_v3.py", line 774, in _step
    async for chunk in invocation:
  File "/app/letta/adapters/simple_llm_stream_adapter.py", line 141, in invoke_llm
    raise self.llm_client.handle_llm_error(e)
letta.errors.LLMError: Unhandled LLM error: Invalid port: 'http:'
Letta.letta.services.streaming_service - ERROR - Run run-227feb5f-270e-47ff-80ff-f9ca83d3494d stopped with LLM error: Unhandled LLM error: Invalid port: 'http:', error_data: {'message_type': 'error_message', 'run_id': 'run-227feb5f-270e-47ff-80ff-f9ca83d3494d', 'error_type': 'llm_error', 'message': 'An error occurred with the LLM request.', 'detail': "Unhandled LLM error: Invalid port: 'http:'", 'seq_id': None}
Letta.letta.services.run_manager - WARNING - Run run-227feb5f-270e-47ff-80ff-f9ca83d3494d completed without a completed_at timestamp
Letta.letta.services.run_manager - WARNING - Run run-227feb5f-270e-47ff-80ff-f9ca83d3494d completed without a completed_at timestamp
Letta.letta.orm.sqlalchemy_base - WARNING - SECURITY: Listing org-scoped model Step without actor. This bypasses organization filtering.
Letta.letta.orm.sqlalchemy_base - WARNING - SECURITY: Listing org-scoped model Step without actor. This bypasses organization filtering.
Letta.letta.server.rest_api.redis_stream_manager - WARNING - [Stream Finalizer] Appending forced [DONE] for run=run-227feb5f-270e-47ff-80ff-f9ca83d3494d (saw_error=True, saw_done=False, final_stop_reason=llm_api_error)
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - WARNING - Context token estimate is not set
Letta.letta.orm.sqlalchemy_base - WARNING - SECURITY: Listing org-scoped model Step without actor. This bypasses organization filtering.
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - WARNING - Error during step processing: Unhandled LLM error: Invalid port: 'http:'
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - INFO - Running final update. Step Progression: StepProgression.START
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - WARNING - Error during agent stream: Unhandled LLM error: Invalid port: 'http:'
Traceback (most recent call last):
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urlparse.py", line 409, in normalize_port
    port_as_int = int(port)
                  ^^^^^^^^^
ValueError: invalid literal for int() with base 10: 'http:'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/app/letta/adapters/simple_llm_stream_adapter.py", line 139, in invoke_llm
    stream = await self.llm_client.stream_async(request_data, self.llm_config)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/letta/otel/tracing.py", line 393, in async_wrapper
    return await func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/letta/llm_api/openai_client.py", line 793, in stream_async
    client = AsyncOpenAI(**kwargs)
             ^^^^^^^^^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/openai/_client.py", line 517, in __init__
    super().__init__(
  File "/app/.venv/lib/python3.11/site-packages/openai/_base_client.py", line 1399, in __init__
    super().__init__(
  File "/app/.venv/lib/python3.11/site-packages/openai/_base_client.py", line 383, in __init__
    self._base_url = self._enforce_trailing_slash(URL(base_url))
                                                  ^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urls.py", line 117, in __init__
    self._uri_reference = urlparse(url, **kwargs)
                          ^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urlparse.py", line 321, in urlparse
    parsed_port: int | None = normalize_port(port, scheme)
                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urlparse.py", line 411, in normalize_port
    raise InvalidURL(f"Invalid port: {port!r}")
httpx.InvalidURL: Invalid port: 'http:'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/app/letta/agents/letta_agent_v3.py", line 377, in stream
    async for chunk in response:
  File "/app/letta/agents/letta_agent_v3.py", line 960, in _step
    raise e
  File "/app/letta/agents/letta_agent_v3.py", line 786, in _step
    raise e
  File "/app/letta/agents/letta_agent_v3.py", line 774, in _step
    async for chunk in invocation:
  File "/app/letta/adapters/simple_llm_stream_adapter.py", line 141, in invoke_llm
    raise self.llm_client.handle_llm_error(e)
letta.errors.LLMError: Unhandled LLM error: Invalid port: 'http:'
Letta.letta.services.streaming_service - ERROR - Run run-787560bf-792c-4bad-97ab-417faa5e9b69 stopped with LLM error: Unhandled LLM error: Invalid port: 'http:', error_data: {'message_type': 'error_message', 'run_id': 'run-787560bf-792c-4bad-97ab-417faa5e9b69', 'error_type': 'llm_error', 'message': 'An error occurred with the LLM request.', 'detail': "Unhandled LLM error: Invalid port: 'http:'", 'seq_id': None}
Letta.letta.services.run_manager - WARNING - Run run-787560bf-792c-4bad-97ab-417faa5e9b69 completed without a completed_at timestamp
Letta.letta.services.run_manager - WARNING - Run run-787560bf-792c-4bad-97ab-417faa5e9b69 completed without a completed_at timestamp
Letta.letta.orm.sqlalchemy_base - WARNING - SECURITY: Listing org-scoped model Step without actor. This bypasses organization filtering.
Letta.letta.orm.sqlalchemy_base - WARNING - SECURITY: Listing org-scoped model Step without actor. This bypasses organization filtering.
Letta.letta.server.rest_api.redis_stream_manager - WARNING - [Stream Finalizer] Appending forced [DONE] for run=run-787560bf-792c-4bad-97ab-417faa5e9b69 (saw_error=True, saw_done=False, final_stop_reason=llm_api_error)
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - WARNING - Context token estimate is not set
Letta.letta.orm.sqlalchemy_base - WARNING - SECURITY: Listing org-scoped model Step without actor. This bypasses organization filtering.
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - WARNING - Error during step processing: Unhandled LLM error: Invalid port: 'http:'
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - INFO - Running final update. Step Progression: StepProgression.START
Letta.agent-fd44f092-caaf-45c5-b76e-bf63e4980701 - WARNING - Error during agent stream: Unhandled LLM error: Invalid port: 'http:'
Traceback (most recent call last):
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urlparse.py", line 409, in normalize_port
    port_as_int = int(port)
                  ^^^^^^^^^
ValueError: invalid literal for int() with base 10: 'http:'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/app/letta/adapters/simple_llm_stream_adapter.py", line 139, in invoke_llm
    stream = await self.llm_client.stream_async(request_data, self.llm_config)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/letta/otel/tracing.py", line 393, in async_wrapper
    return await func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/letta/llm_api/openai_client.py", line 793, in stream_async
    client = AsyncOpenAI(**kwargs)
             ^^^^^^^^^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/openai/_client.py", line 517, in __init__
    super().__init__(
  File "/app/.venv/lib/python3.11/site-packages/openai/_base_client.py", line 1399, in __init__
    super().__init__(
  File "/app/.venv/lib/python3.11/site-packages/openai/_base_client.py", line 383, in __init__
    self._base_url = self._enforce_trailing_slash(URL(base_url))
                                                  ^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urls.py", line 117, in __init__
    self._uri_reference = urlparse(url, **kwargs)
                          ^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urlparse.py", line 321, in urlparse
    parsed_port: int | None = normalize_port(port, scheme)
                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.11/site-packages/httpx/_urlparse.py", line 411, in normalize_port
    raise InvalidURL(f"Invalid port: {port!r}")
httpx.InvalidURL: Invalid port: 'http:'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/app/letta/agents/letta_agent_v3.py", line 377, in stream
    async for chunk in response:
  File "/app/letta/agents/letta_agent_v3.py", line 960, in _step
    raise e
  File "/app/letta/agents/letta_agent_v3.py", line 786, in _step
    raise e
  File "/app/letta/agents/letta_agent_v3.py", line 774, in _step
    async for chunk in invocation:
  File "/app/letta/adapters/simple_llm_stream_adapter.py", line 141, in invoke_llm
    raise self.llm_client.handle_llm_error(e)
letta.errors.LLMError: Unhandled LLM error: Invalid port: 'http:'
Letta.letta.services.streaming_service - ERROR - Run run-f2c55435-eaa9-4b5d-b2f9-30f7a9ca052b stopped with LLM error: Unhandled LLM error: Invalid port: 'http:', error_data: {'message_type': 'error_message', 'run_id': 'run-f2c55435-eaa9-4b5d-b2f9-30f7a9ca052b', 'error_type': 'llm_error', 'message': 'An error occurred with the LLM request.', 'detail': "Unhandled LLM error: Invalid port: 'http:'", 'seq_id': None}
Letta.letta.services.run_manager - WARNING - Run run-f2c55435-eaa9-4b5d-b2f9-30f7a9ca052b completed without a completed_at timestamp
Letta.letta.services.run_manager - ERROR - Run run-f2c55435-eaa9-4b5d-b2f9-30f7a9ca052b is already in a terminal state failed with stop reason llm_api_error, but is being updated with data {'status': <RunStatus.failed: 'failed'>, 'completed_at': None, 'stop_reason': <StopReasonType.llm_api_error: 'llm_api_error'>, 'metadata': {'error': {'message_type': 'error_message', 'run_id': 'run-f2c55435-eaa9-4b5d-b2f9-30f7a9ca052b', 'error_type': 'llm_error', 'message': 'An error occurred with the LLM request.', 'detail': "Unhandled LLM error: Invalid port: 'http:'", 'seq_id': None}}, 'total_duration_ns': None}
Letta.letta.orm.sqlalchemy_base - WARNING - SECURITY: Listing org-scoped model Step without actor. This bypasses organization filtering.
Letta.letta.orm.sqlalchemy_base - WARNING - SECURITY: Listing org-scoped model Step without actor. This bypasses organization filtering.
Letta.letta.server.rest_api.redis_stream_manager - WARNING - [Stream Finalizer] Appending forced [DONE] for run=run-f2c55435-eaa9-4b5d-b2f9-30f7a9ca052b (saw_error=True, saw_done=False, final_stop_reason=llm_api_error)

Observations

Server log indicates that an OpenAI client was initialized.
This is unexpected behavior since I only enable Ollama provider on my local Letta server.

[cpacker]

For testing purposes, can you try running Ollama on the default port and following the commands exactly? https://docs.letta.com/letta-code/docker#prerequisites

[ansidev]

@cpacker To be clear, OLLAMA_PORT=11434.

docker exec -it letta-server bash

Then run

root@bc6df345af9c:/app# env
HOSTNAME=bc6df345af9c
POSTGRES_PASSWORD=letta
PWD=/app
LETTA_ENVIRONMENT=DEV
LETTA_VERSION=
OLLAMA_BASE_URL=http://host.docker.internal:11434/v1
HOME=/root
LANG=en_US.utf8
VIRTUAL_ENV=/app/.venv
GOSU_VERSION=1.16
PG_MAJOR=15
PG_VERSION=15.4-2.pgdg120+1
TERM=xterm
SHLVL=1
POSTGRES_USER=letta
PGDATA=/var/lib/postgresql/data
PATH=/app/.venv/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/lib/postgresql/15/bin
POSTGRES_DB=letta
_=/usr/bin/env

[cpacker]

Right I mean can you just try the actual command verbatim in case there's some weird string mangling happening

[ansidev]

@cpacker To be clear, I'm not convinced by your suggestion (since the environment variable is correct). However, I did that, and the issue still remains (of course).

From my perspective, we should be able to debug the request submitted by the client to the server, then we can narrow down whether the issue came from the CLI or the server.

If you can provide any guidances, that would be helpful.

[ansidev]

As I mentioned, I verified that Ollama models are updated in letta CLI using /model.
Therefore, I believe it means some requests were submitted correctly to the letta server.

However, the Ollama provider does not display in the output of the /connect command. Is it relevant?

[cpacker]

/connect is not for ollama, that's for BYOK w/ Letta API: https://docs.letta.com/letta-code/providers

Ollama docs are here: https://docs.letta.com/letta-code/docker

I'm guessing this is a server bug, try potentially dropping the /v1, or also connect Letta Code to a good model (eg Sonnet/Opus/GPT-5.2) or use your coding agent of choice, then point it at the running docker image and ask it to look inside at the source code to trace the issue - probably is fixable w/ a one-line change to something in the server

[cpacker]

(will also take a look at this once I have time, but follow the above if you need to unblock yourself asap)

[ansidev]

Is there any way we can run a debugable version of the Letta server?

[cpacker]

https://github.com/letta-ai/letta?tab=contributing-ov-file#-install-dependencies--configure-environment