Skip to content

fix(v1): preserve cache usage in train responses #2542

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

Draft
hallerite wants to merge 1 commit into
base: fix/eval-long-override-display
Choose a base branch
from
Draft

fix(v1): preserve cache usage in train responses #2542

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
9 changes: 6 additions & 3 deletions verifiers/v1/clients/train.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,7 @@
from typing import Any, ClassVar, TypeVar

from openai import OpenAIError
from openai.types import CompletionUsage
from renderers import OverlongPromptError, RenderedTokens, Renderer, RendererConfig
from renderers.base import ToolCallParseStatus, is_multimodal

Expand Down Expand Up @@ -145,9 +146,11 @@ def response_from_generate(
tool_calls=tool_calls,
),
finish_reason=finish,
# /inference/v1/generate returns exact token ids but no usage details, so the
# completion's reasoning-token subset is unknown.
usage=Usage(
# Preserve provider cache/reasoning details through the same accounting as chat.
# Older endpoints/renderers may omit usage; exact token lengths remain a fallback.
usage=Usage.from_openai(CompletionUsage.model_validate(result["usage"]))
if result.get("usage") is not None
else Usage(
prompt_tokens=len(prompt_ids), completion_tokens=len(completion_ids)
),
# generate() returns owned, typed lists. Skip revalidation here to avoid copying
Expand Down