Verification register Compute & Web Infra
Current reading
The readiness gap, in one scan
AI-assisted assembly · derived results
- Claimed
- 90
- Reported
- 86
- Verified
- 77
- Gap
- +13
Public ambition and stated capability
Observed practitioner reporting
Independently supported evidence
Claimed minus verified
Evidence strength Strong
Decision
What the current evidence supports
Human editorial judgment · 2026-06-28
Adopt with guardrails
- Why
- Mature, Apache-2.0, very-high activity (29.6k stars, weekly releases, xAI/Cursor/cloud adopters) and competitive performance — a safe pick for self-hosted serving, but adoption is conditional on a GPU-hosting use case we don't currently have.
- Next
- Keep as the default open-source engine of record IF/when we self-host LLM inference on GPUs; benchmark SGLang vs TensorRT-LLM on the specific target model+workload before committing, since prefix-sharing determines whether RadixAttention pays off.
Constraints
Blockers
No named blocker is present in the current public projection.
Evidence summary
Derived counts
AI-assisted assembly
- Total
- 8
- Tier 1
- 2
- Tier 2
- 3
- Tier 3
- 3
- Supports
- 5
- Contradicts
- 3
- Context
- 0
- Latest observed
- 2026-06-13
Counts and dates only. Raw signals, private excerpts, trust records, and internal corpus material are not published here.
Publication record
Revisions
Initial public reading
This is the initial public reading. No earlier readiness change is recorded.