fix(vllm): initializes dictionary entry prior to usage

[CollectiveUnicorn]

Description

Fixes exception occurring when the dictionary entry has not been instantiated before being referenced.

Exception:

│ leapfrogai-model Traceback (most recent call last):                                                                                                                                                                                                                                                                                                                    │
│ leapfrogai-model   File "/home/leapfrogai/.pyenv/versions/3.11.6/lib/python3.11/asyncio/events.py", line 80, in _run                                                                                                                                                                                                                                                   │
│ leapfrogai-model     self._context.run(self._callback, *self._args)                                                                                                                                                                                                                                                                                                    │
│ leapfrogai-model   File "/home/leapfrogai/.venv/lib/python3.11/site-packages/vllm/engine/async_llm_engine.py", line 38, in _raise_exception_on_finish                                                                                                                                                                                                                  │
│ leapfrogai-model     task.result()                                                                                                                                                                                                                                                                                                                                     │
│ leapfrogai-model   File "/home/leapfrogai/.venv/lib/python3.11/site-packages/vllm/engine/async_llm_engine.py", line 508, in run_engine_loop                                                                                                                                                                                                                            │
│ leapfrogai-model     await asyncio.sleep(0)                                                                                                                                                                                                                                                                                                                            │
│ leapfrogai-model   File "/home/leapfrogai/.pyenv/versions/3.11.6/lib/python3.11/asyncio/tasks.py", line 640, in sleep                                                                                                                                                                                                                                                  │
│ leapfrogai-model     await __sleep0()                                                                                                                                                                                                                                                                                                                                  │
│ leapfrogai-model   File "/home/leapfrogai/.pyenv/versions/3.11.6/lib/python3.11/asyncio/tasks.py", line 634, in __sleep0                                                                                                                                                                                                                                               │
│ leapfrogai-model     yield                                                                                                                                                                                                                                                                                                                                             │
│ leapfrogai-model asyncio.exceptions.CancelledError                                                                                                                                                                                                                                                                                                                     │
│ leapfrogai-model Exception in thread Thread-2 (run):                                                                                                                                                                                                                                                                                                                   │
│ leapfrogai-model Traceback (most recent call last):                                                                                                                                                                                                                                                                                                                    │
│ leapfrogai-model   File "/home/leapfrogai/.pyenv/versions/3.11.6/lib/python3.11/threading.py", line 1045, in _bootstrap_inner                                                                                                                                                                                                                                          │
│ leapfrogai-model     self.run()                                                                                                                                                                                                                                                                                                                                        │
│ leapfrogai-model   File "/home/leapfrogai/.pyenv/versions/3.11.6/lib/python3.11/threading.py", line 982, in run                                                                                                                                                                                                                                                        │
│ leapfrogai-model     self._target(*self._args, **self._kwargs)                                                                                                                                                                                                                                                                                                         │
│ leapfrogai-model   File "/home/leapfrogai/.pyenv/versions/3.11.6/lib/python3.11/asyncio/runners.py", line 190, in run                                                                                                                                                                                                                                                  │
│ leapfrogai-model     return runner.run(main)                                                                                                                                                                                                                                                                                                                           │
│ leapfrogai-model            ^^^^^^^^^^^^^^^^                                                                                                                                                                                                                                                                                                                           │
│ leapfrogai-model   File "/home/leapfrogai/.pyenv/versions/3.11.6/lib/python3.11/asyncio/runners.py", line 118, in run                                                                                                                                                                                                                                                  │
│ leapfrogai-model     return self._loop.run_until_complete(task)                                                                                                                                                                                                                                                                                                        │
│ leapfrogai-model            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^                                                                                                                                                                                                                                                                                                        │
│ leapfrogai-model   File "/home/leapfrogai/.pyenv/versions/3.11.6/lib/python3.11/asyncio/base_events.py", line 653, in run_until_complete                                                                                                                                                                                                                               │
│ leapfrogai-model     return future.result()                                                                                                                                                                                                                                                                                                                            │
│ leapfrogai-model            ^^^^^^^^^^^^^^^                                                                                                                                                                                                                                                                                                                            │
│ leapfrogai-model   File "/home/leapfrogai/src/main.py", line 184, in iterate_outputs                                                                                                                                                                                                                                                                                   │
│ leapfrogai-model     f"Generated {num_tokens_by_id[request_id]} tokens in {time.time() - t0_by_id[request_id]:.2f}s"                                                                                                                                                                                                                                                   │
│ leapfrogai-model                  ~~~~~~~~~~~~~~~~^^^^^^^^^^^^                                                                                                                                                                                                                                                                                                         │
│ leapfrogai-model KeyError: 'e6271f2482fb48a181a2ee79c871b3df'

BREAKING CHANGES

CHANGES

• Prevents vllm from reaching a finished state without first initializing the token count and text delta.

Related Issue

Fixes #

Relates to #

Checklist before merging

• Tests, documentation, ADR added or updated as needed
• Followed the Contributor Guide Steps

[netlify]

✅ Deploy Preview for leapfrogai-docs canceled.

Name	Link
🔨 Latest commit	462ae30
🔍 Latest deploy log	https://app.netlify.com/sites/leapfrogai-docs/deploys/66cf65e009fafc00081c4d18