[Bug]: `vision_analyze` on a slow enough local model will timeout due to default 30s timout

### Bug Description

`vision_analyze` fails with a timeout error when using a local model for the auxiliary vision client. The default 30s timeout in `async_call_llm` is too short for local models where image encoding alone can take 30s+ to complete (on very constrained hardware or large/slow models on consumer hardware etc)

### Steps to Reproduce

1. Use locally hosted model with vision
2. Ask hermes to describe an image
3. Wait 30~ seconds for timeout to occur

### Expected Behavior

Image is analyzed successfully and the result is passed to the main model

### Actual Behavior

Vision pre-analysis times out before the server returns the first token:

```
httpx.ReadTimeout
openai.APITimeoutError: Request timed out.
ERROR tools.vision_tools: Error analyzing image: Request timed out.
```

### Affected Component

Tools (terminal, file ops, web, code execution, etc.)

### Messaging Platform (if gateway-related)

_No response_

### Operating System

Arch Linux (6.19.7-1)

### Python Version

3.11.15

### Hermes Version

v0.4.0 (2026.3.18)

### Relevant Logs / Traceback

<details>
<summary>Expand: ERROR tools.vision_tools: Error analyzing image: Request timed out</summary>

```shell
2026-03-19 20:08:23,828 ERROR tools.vision_tools: Error analyzing image: Request timed out.
Traceback (most recent call last):
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpx/_transports/default.py", line 101, in map_httpcore_exceptions
    yield
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpx/_transports/default.py", line 394, in handle_async_request
    resp = await self._pool.handle_async_request(req)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpcore/_async/connection_pool.py", line 256, in handle_async_request
    raise exc from None
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpcore/_async/connection_pool.py", line 236, in handle_async_request
    response = await connection.handle_async_request(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpcore/_async/connection.py", line 103, in handle_async_request
    return await self._connection.handle_async_request(request)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpcore/_async/http11.py", line 136, in handle_async_request
    raise exc
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpcore/_async/http11.py", line 106, in handle_async_request
    ) = await self._receive_response_headers(**kwargs)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpcore/_async/http11.py", line 177, in _receive_response_headers
    event = await self._receive_event(timeout=timeout)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpcore/_async/http11.py", line 217, in _receive_event
    data = await self._network_stream.read(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpcore/_backends/anyio.py", line 32, in read
    with map_exceptions(exc_map):
  File "/home/smt/.local/share/uv/python/cpython-3.11.15-linux-x86_64-gnu/lib/python3.11/contextlib.py", line 158, in __exit__
    self.gen.throw(typ, value, traceback)
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpcore/_exceptions.py", line 14, in map_exceptions
    raise to_exc(exc) from exc
httpcore.ReadTimeout

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/openai/_base_client.py", line 1604, in request
    response = await self._client.send(
               ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpx/_client.py", line 1629, in send
    response = await self._send_handling_auth(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpx/_client.py", line 1657, in _send_handling_auth
    response = await self._send_handling_redirects(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpx/_client.py", line 1694, in _send_handling_redirects
    response = await self._send_single_request(request)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpx/_client.py", line 1730, in _send_single_request
    response = await transport.handle_async_request(request)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpx/_transports/default.py", line 393, in handle_async_request
    with map_httpcore_exceptions():
  File "/home/smt/.local/share/uv/python/cpython-3.11.15-linux-x86_64-gnu/lib/python3.11/contextlib.py", line 158, in __exit__
    self.gen.throw(typ, value, traceback)
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/httpx/_transports/default.py", line 118, in map_httpcore_exceptions
    raise mapped_exc(message) from exc
httpx.ReadTimeout

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/home/smt/.hermes/hermes-agent/tools/vision_tools.py", line 310, in vision_analyze_tool
    response = await async_call_llm(**call_kwargs)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/agent/auxiliary_client.py", line 1512, in async_call_llm
    return await client.chat.completions.create(**kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/openai/resources/chat/completions/completions.py", line 2714, in create
    return await self._post(
           ^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/openai/_base_client.py", line 1884, in post
    return await self.request(cast_to, opts, stream=stream, stream_cls=stream_cls)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/smt/.hermes/hermes-agent/venv/lib/python3.11/site-packages/openai/_base_client.py", line 1622, in request
    raise APITimeoutError(request=request) from err
openai.APITimeoutError: Request timed out.
```

</details>

### Root Cause Analysis (optional)

inside of `vision_analyze_tool()` the following call_kwargs is defined:

```python
call_kwargs = {
            "task": "vision",
            "messages": messages,
            "temperature": 0.1,
            "max_tokens": 2000,
        }
```

without a timeout, the timeout ends up defaulting somewhere up the chain, many of them like `async_call_llm()` in `agent/auxiliary_client.py` have a default of 30

as a quick workaround I was able to add `"timeout": 180,` to call_kwargs which stopped the timeout

### Proposed Fix (optional)

some kind of configuration or envvar would be useful as just bumping the default could obviously lead to some unwanted behavior for people who *know* their models should be returning fast, but for people playing around with local models and constrained hardware, some way to raise it would be useful

potentially also a more generous default if the provider is detected as custom/local would give a smoother experience for people using a wider range of configurations

### Are you willing to submit a PR for this?

- [ ] I'd like to fix this myself and submit a PR

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[Bug]: `vision_analyze` on a slow enough local model will timeout due to default 30s timout #2107

Bug Description

Steps to Reproduce

Expected Behavior

Actual Behavior

Affected Component

Messaging Platform (if gateway-related)

Operating System

Python Version

Hermes Version

Relevant Logs / Traceback

Root Cause Analysis (optional)

Proposed Fix (optional)

Are you willing to submit a PR for this?

Metadata

Assignees

Labels

Type

Fields

Projects

Milestone

Relationships

Development

[Bug]: vision_analyze on a slow enough local model will timeout due to default 30s timout #2107

Description

Bug Description

Steps to Reproduce

Expected Behavior

Actual Behavior

Affected Component

Messaging Platform (if gateway-related)

Operating System

Python Version

Hermes Version

Relevant Logs / Traceback

Root Cause Analysis (optional)

Proposed Fix (optional)

Are you willing to submit a PR for this?

Metadata

Metadata

Assignees

Labels

Type

Fields

Projects

Milestone

Relationships

Development

Issue actions

[Bug]: `vision_analyze` on a slow enough local model will timeout due to default 30s timout #2107