Thanks to visit codestin.com
Credit goes to github.com

Skip to content

feat: update interpreter to print out model response messages in async mode #717

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Mar 7, 2025
Merged

feat: update interpreter to print out model response messages in async mode #717

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
25 changes: 15 additions & 10 deletions src/pdl/pdl_llms.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -87,7 +87,7 @@ def generate_text(
messages: ModelInput,
parameters: dict[str, Any],
) -> tuple[LazyMessage, PdlLazy[Any]]:
print("Asynchronous model call started", file=stderr)
print(f"Asynchronous model call started to {block.model}", file=stderr)
# global _BACKGROUND_TASKS
future = asyncio.run_coroutine_threadsafe(
LitellmModel.async_generate_text(
Expand All @@ -110,24 +110,29 @@ def update_end_nanos(future):
if block.pdl__timing is not None:
block.pdl__timing.end_nanos = time.time_ns()

# report call completion and its duration (and if available queueing time)
res = future.result()[1]
# report call completion and its duration
start = (
block.pdl__timing.start_nanos
if block.pdl__timing.start_nanos is not None
else 0
)
exec_nanos = block.pdl__timing.end_nanos - start
queue_nanos = 0
if "created" in res:
queue_nanos = (
res["created"] * 1000000000 - block.pdl__timing.start_nanos
)
exec_nanos = exec_nanos - queue_nanos
print(
f"Asynchronous model call to {block.model} completed in {(exec_nanos)/1000000}ms queued for {queue_nanos/1000000}ms",
f"Asynchronous model call to {block.model} completed in {(exec_nanos)/1000000}ms",
file=stderr,
)
msg = future.result()[0]
if msg["content"] is not None:
from termcolor import colored

from .pdl_ast import BlockKind
from .pdl_scheduler import color_of

print(
colored(msg["content"], color=color_of(BlockKind.MODEL)),
file=stderr,
)
print("\n", file=stderr)

future.add_done_callback(update_end_nanos)

Expand Down