scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-07 08:26:10 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 36

docs: add extra operations to mermaid diagram

Browse source 
Co-authored-by: ishaan <ishaan@berri.ai>
This commit is contained in:
Cursor Agent 2026-01-23 19:16:47 +00:00
parent 108ed009fd
commit bf36fa7652
1 changed files with 16 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

23
docs/my-website/docs/pass_through/architecture.md
View file

@ -16,23 +16,33 @@ sequenceDiagram
rect rgb(240, 240, 240)
Note over Proxy: 1. URL Construction
Note over Proxy: Build: https://us-central1-aiplatform.googleapis.com/...
Note over Proxy: Build regional/provider-specific URL
end
rect rgb(240, 240, 240)
Note over Proxy: 2. Auth Header Replacement
Note over Proxy: Replace litellm key → provider credentials
Note over Proxy: LiteLLM key → provider credentials
end
rect rgb(240, 240, 240)
Note over Proxy: 3. Extra Operations
Note over Proxy: • x-pass-* headers (strip prefix, forward)
Note over Proxy: • x-litellm-tags → metadata
Note over Proxy: • Guardrails (opt-in)
Note over Proxy: • Multipart form reconstruction
end
Proxy->>Provider: POST https://us-central1-aiplatform.googleapis.com/...
Note over Proxy,Provider: Headers: Authorization: Bearer ya29.google-oauth...
Note over Proxy,Provider: Body: { "contents": [...] } ← UNCHANGED
Provider-->>Proxy: Response
Provider-->>Proxy: Response (streaming or non-streaming)
rect rgb(240, 240, 240)
Note over Proxy: 3. Logging (async, optional)
Note over Proxy: Parse response → calculate cost → log
Note over Proxy: 4. Response Handling (async)
Note over Proxy: • Collect streaming chunks for logging
Note over Proxy: • Cost injection (if enabled)
Note over Proxy: • Parse response → calculate cost → log
end
Proxy-->>Client: Response (unchanged)
@ -42,7 +52,6 @@ sequenceDiagram
- **URL Construction** - Build correct provider URL (e.g., regional endpoints for Vertex AI, Bedrock)
- **Auth Header Replacement** - Swap LiteLLM virtual key for actual provider credentials
- **Logging** (optional) - Parse response to extract usage and calculate cost
## Extra Operations