Skip to content

Serialize intercepted completions directly to JSON bytes #1795

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
xeophon merged 1 commit into from
Jun 21, 2026
Merged

Serialize intercepted completions directly to JSON bytes #1795

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
17 changes: 13 additions & 4 deletions verifiers/v1/interception/server.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,7 @@

from aiohttp import web
from pydantic import ValidationError
from pydantic_core import from_json
from pydantic_core import PydanticSerializationError, from_json, to_json

from verifiers.v1.clients import RolloutContext
from verifiers.v1.dialects import DIALECTS, Dialect
Expand Down Expand Up @@ -60,6 +60,15 @@
_HOST = "127.0.0.1"


def _completion_response(completion: dict | None) -> web.Response:
"""Serialize a model's JSON-native response without an intermediate string."""
try:
body = to_json(completion, inf_nan_mode="constants")
except PydanticSerializationError:
return web.json_response(completion)
return web.Response(body=body, content_type="application/json", charset="utf-8")


@dataclass(frozen=True)
class RolloutLimits:
"""Per-rollout framework limits (None = no cap), checked before each turn is served.
Expand Down Expand Up @@ -292,7 +301,7 @@ async def handle_request(
return web.json_response(
dialect.error_body(f"rollout stopped: {refused}"), status=400
)
return web.json_response(completion)
return _completion_response(completion)
turn = graph.prepare_turn(session.trace, prompt)
session.error = None
try:
Expand All @@ -316,7 +325,7 @@ async def handle_request(
dialect.error_body("rollout stopped: context_length"),
status=400,
)
return web.json_response(completion)
return _completion_response(completion)
except RolloutError as e:
# Stash the real cause; the rollout re-raises it after the harness returns. Relay
# the provider's status so the harness SDK retries 5xx/429 and not 4xx.
Expand Down Expand Up @@ -351,7 +360,7 @@ async def handle_request(
# Hand back to the program when the model wants a tool (the program runs it) or
# when there's no user simulator to keep the conversation going.
if response.message.tool_calls or session.user is None:
return web.json_response(completion)
return _completion_response(completion)
try:
user_messages = await session.user(response.message.content or "")
except RolloutError as e:
Expand Down
Loading