docs(readme): align README with the shipped feature set - #2030
Conversation
|
Caution Review failedThe pull request is closed. ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (1)
📝 WalkthroughSummary by CodeRabbit
WalkthroughThe README now describes expanded gateway capabilities, APIs, routing policies, supported backends, installation methods, prerequisites, documentation, and contribution resources. ChangesREADME documentation refresh
Estimated code review effort: 1 (Trivial) | ~5 minutes Suggested reviewers: 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
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. Comment |
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@README.md`:
- Line 37: Update the Full Observability entry in README.md to use the shipped
Prometheus metric count that matches the hosted documentation, changing the
inaccurate “90+” claim to “40+” unless the documentation and metric inventory
are updated together.
- Line 101: Reconcile the parser counts shown in the README’s Model Parsers
entry with the authoritative hosted gRPC parser inventory. Update the README
counts or the hosted inventory so both list the same shipped tool-call and
reasoning parser totals, while preserving the existing model-detection
description.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
8035cc6 to
876b8c9
Compare
Every claim now traces to code or the supported product surface: - 8 -> 10 routing policies (adds passthrough, least_load) - 40+ -> 90+ Prometheus metrics; OTel W3C trace propagation into engines over HTTP and gRPC - Drop unverifiable claims: sub-millisecond routing decisions, zero-copy tokenization (now: cached tokenization with zero-copy cache hits) - Supported Backends reshaped to one row per category so the two lists no longer read as row pairings; engines gain MLX, Ollama reclassified as an OpenAI-compatible server; providers list the dedicated adapters (OpenAI, Anthropic, Gemini, xAI) followed by OCI Generative AI, AWS Bedrock, and Azure OpenAI (reached via their OpenAI-compatible modes) and the generic passthrough - API coverage now lists the shipped surface: Rerank, Classify, Conversations, Gemini Interactions, Realtime (WebSocket + WebRTC), audio transcription, tokenize/detokenize; MCP execution scoped to Responses + Messages where it actually runs - New Why/Features rows: priority admission scheduling, PD/EPD disaggregation with DP-aware routing, K8s discovery, 21 tool + 16 reasoning parsers; OIDC correctly scoped to the control plane - Quick Start: add Helm (oci://ghcr.io/smg-project/charts/smg) and the protoc prerequisite for cargo install - All docs links moved from smg-docs GitHub blobs to the live site (lightseek.org/smg); docs badge retargeted from the dead MkDocs GitHub Pages site; Docker badge now points at Docker Hub where stable tags are actually published Signed-off-by: Simo Lin <25425177+slin1237@users.noreply.github.com>
876b8c9 to
c13c327
Compare
Description
Problem
The README had drifted from the codebase in both directions — overstating some things (unverifiable performance claims, a self-contradicting provider story) and underselling others (roughly half the shipped API surface was unmentioned). Doc links pointed at raw markdown blobs in the smg-docs repo instead of the live site, and two badges pointed at the wrong places. The Supported Backends table also rendered two unrelated lists side by side, which read as row pairings ("SGLang ↔ OpenAI").
Solution
Full accuracy pass, verified claim-by-claim against the code (four parallel code audits; file:line evidence for each item below), plus a reshaped backends table.
Corrected (was overstated or wrong):
passthrough,least_loadwere missing; policy factory inmodel_gateway/src/policies/factory.rs)Arc'd token IDs; encoding itself is not zero-copy)Supported Backends reshaped — one row per category so the two lists no longer read as pairings:
Added (shipped but undocumented):
oci://ghcr.io/smg-project/charts/smg) and theprotocprerequisite forcargo install smgLinks & badges:
github.com/smg-project/smg-docs/blob/...to the live site (https://lightseek.org/smg/...) — every target curl-verified 200lightseek.org/smglightseekorg/smg) with a live version shield — stable tags are only published there; ghcr.io/lightseekorg gets nightlies onlyDeliberately not claimed: ZMQ connection mode (crate + detection exist, but the request path is still gRPC-only per
routers/grpc/backend_client.rs).Test Plan
CliArgs(launchsubcommand,--worker-urls,--policy cache_aware, mesh flags, default port 30000)Checklist
cargo +nightly fmtpassescargo clippy --all-targets --all-features -- -D warningspasses (not runnable on this machine — missing OpenCV system deps; no Rust touched, CI is authoritative)