fix(anthropic): stop double-counting output tokens on streamed messages

[arthi-arumugam-git]

Problem

_process_response_item treats message_delta.usage.output_tokens as an increment and adds it to the count already stored from message_start:

item_output_tokens = dict(item.usage).get("output_tokens", 0)
existing_output_tokens = complete_response["usage"].get("output_tokens", 0)
complete_response["usage"]["output_tokens"] = (
    item_output_tokens + existing_output_tokens
)

Anthropic sends message_delta.usage.output_tokens as the running total for the whole message. The value in message_start is a partial count that is already contained in it. Adding them inflates output tokens, and therefore total tokens and any cost computed downstream, on every streamed Anthropic span. Nothing raises, so the wrong number just propagates.

The vendor SDK assigns rather than adds, in anthropic/lib/streaming/_messages.py:

current_snapshot.usage.output_tokens = event.usage.output_tokens

This repo's own Bedrock streaming wrapper also overwrites rather than accumulates (bedrock/streaming_wrapper.py, usage.update(...)).

Reproduction

Replaying the SSE events from this repo's existing cassette, tests/cassettes/test_messages/test_anthropic_message_streaming_legacy.yaml, through both accumulators:

message_start.usage : {'input_tokens': 17, 'output_tokens': 3}
message_delta.usage : {'output_tokens': 171}

openllmetry  output_tokens : 174
anthropic SDK output_tokens: 171   <- ground truth

A 1.75% overcount on this recording. The gap is whatever message_start reported, so it grows with prompt-cache and tool-use shapes where that initial count is larger.

Fix

Overwrite with the delta's value instead of adding, and leave the stored count alone when the delta omits output_tokens.

Test

test_anthropic_message_streaming_legacy already asserted input_tokens == 17 and the input + output == total identity, but never asserted the output value itself, so the bug passed CI. This PR adds that one assertion. It fails with assert 174 == 171 on main and passes with the fix.

Summary by CodeRabbit

• Bug Fixes
  • Corrected streamed output token reporting to use the provider’s cumulative totals, preventing inflated usage counts.
  • Ensured usage metrics remain accurate when token totals are omitted from individual streaming updates.

[CLAassistant]

All committers have signed the CLA.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 7a5f5fc2-bc20-48a2-8280-61ab4d6983d6

📥 Commits

Reviewing files that changed from the base of the PR and between 93429cf and b3722db.

📒 Files selected for processing (2)

• packages/opentelemetry-instrumentation-anthropic/opentelemetry/instrumentation/anthropic/streaming.py
• packages/opentelemetry-instrumentation-anthropic/tests/test_messages.py

---

📝 Walkthrough

Walkthrough

Anthropic streaming usage handling now treats output_tokens as an upstream running total and verifies the resulting span value in the legacy streaming test.

Changes

Anthropic streaming usage

Layer / File(s)	Summary
Record output token totals packages/opentelemetry-instrumentation-anthropic/.../streaming.py, packages/opentelemetry-instrumentation-anthropic/tests/test_messages.py	message_delta handling overwrites explicitly provided output token totals instead of accumulating them, and the legacy streaming test asserts an output token count of 171.

Estimated code review effort: 2 (Simple) | ~10 minutes

Possibly related PRs

• traceloop/openllmetry#4372: Both changes update Anthropic streaming message_delta handling in streaming.py.

Suggested reviewers: dvirski

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly matches the main change: fixing Anthropic streamed message output-token double counting.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[arthi-arumugam-git]

Gentle ping: CodeRabbit, GitGuardian, and the CLA are all green, and the branch is current as of yesterday. The test workflows are still waiting on the fork workflow approval click; would a maintainer be able to approve the run so CI can report on this change?