docs(sagemaker): distinguish native sagemaker_chat streaming from legacy fake streaming

[devin-ai-integration]

Summary

The SageMaker page said flatly that "Sagemaker currently does not support streaming", which is stale for the sagemaker_chat/ route. That statement is only true for the legacy sagemaker/ completion route (which hits /invocations and fakes streaming by chunking the full string); sagemaker_chat/ hits /invocations-response-stream and streams natively.

Two edits to docs/providers/aws_sagemaker.md:

• the "Usage - Streaming" note now scopes the fake-streaming caveat to the legacy sagemaker/ route and points readers to the Messages API section for native streaming
• the "Sagemaker Messages API" section gains a short paragraph stating that sagemaker_chat/ streams natively, forwards each decoded event as it arrives (so TTFT and chunk cadence track the endpoint), and that stream=True puts "stream": true in the signed request body

Companion to the code fix in BerriAI/litellm#34338 (LIT-4313), which removed the fixed 1024-byte read buffer that was inflating client TTFT on sagemaker_chat streams.

Link to Devin session: https://app.devin.ai/sessions/27eec2ce9b6a4f6db8a8fbc96dd95e39
Requested by: @shivamrawat1

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment, CI, and merge conflict monitoring

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
litellm	Ready	Preview, Comment	Jul 23, 2026 1:20am