WFGY/ProblemMap/GlobalFixMap/LocalDeploy_Inference/textgen-webui.md

TextGen WebUI: Guardrails and Fix Patterns

🧭 Quick Return to Map

You are in a sub-page of LocalDeploy_Inference.
To reorient, go back here:

• LocalDeploy_Inference — on-prem deployment and model inference
• WFGY Global Fix Map — main Emergency Room, 300+ structured fixes
• WFGY Problem Map 1.0 — 16 reproducible failure modes

Think of this page as a desk within a ward.
If you need the full triage and all prescriptions, return to the Emergency Room lobby.

A stabilization guide for oobabooga/text-generation-webui (TextGen WebUI). This tool is popular for local or Docker-based inference but often drifts under plugins, extensions, or mixed model backends. Use this page to detect and repair instability without changing infra.

---

Open these first

• Architecture & recovery: RAG Architecture & Recovery
• Retrieval knobs: Retrieval Playbook
• Trace schema: retrieval-traceability.md
• Ordering & warm-up fixes: bootstrap-ordering.md, deployment-deadlock.md, predeploy-collapse.md
• Embedding check: embedding-vs-semantic.md

---

Core acceptance

• ΔS(question, retrieved) ≤ 0.45
• Coverage ≥ 0.70 to target snippet
• λ remains convergent across 3 paraphrases and 2 seeds
• Stable across UI plugins (no flip-flops when toggling extensions)

---

Typical WebUI breakpoints and fix

Symptom	Likely cause	Fix
Runs fine in notebook, but WebUI returns divergent outputs	Extension modifies prompt schema	data-contracts.md, retrieval-traceability.md
Citations missing when using API mode	WebUI truncates field names	retrieval-traceability.md
ΔS > 0.60 only when multiple users connect	Concurrency split in worker threads	context-drift.md, entropy-collapse.md
JSON schema output fails with extensions	UI prompt injection / malformed tools	prompt-injection.md, logic-collapse.md
First run after model load unstable	Race in init hooks	bootstrap-ordering.md

---

Fix in 60 seconds

1. Warm-up: Always run a dummy inference after model load. Log INDEX_HASH.
2. Schema lock: Enforce snippet_id, tokens, offsets in every response.
3. λ probe: Run 3 paraphrases, 2 seeds. Clamp with BBAM if divergent.
4. Trace logging: Save ΔS and λ per extension toggle.
5. Verify: Replay dataset with and without extensions enabled. ΔS ≤ 0.45 required.

---

Copy-paste diagnostic prompt

I am running text-generation-webui with {extensions_list}.  
Question: "{user_question}"  

Please return:
- ΔS vs retrieved snippet  
- λ across paraphrases  
- Citations preserved? (yes/no)  
- If ΔS ≥ 0.60, point to minimal WFGY module to apply  

---

🔗 Quick-Start Downloads (60 sec)

Tool	Link	3-Step Setup
WFGY 1.0 PDF	Engine Paper	1️⃣ Download · 2️⃣ Upload to your LLM · 3️⃣ Ask “Answer using WFGY + <your question>”
TXT OS (plain-text OS)	TXTOS.txt	1️⃣ Download · 2️⃣ Paste into any LLM chat · 3️⃣ Type “hello world” — OS boots instantly

---

Explore More

Layer	Page	What it’s for
Proof	WFGY Recognition Map	External citations, integrations, and ecosystem proof
Engine	WFGY 1.0	Original PDF based tension engine
Engine	WFGY 2.0	Production tension kernel and math engine for RAG and agents
Engine	WFGY 3.0	TXT based Singularity tension engine, 131 S class set
Map	Problem Map 1.0	Flagship 16 problem RAG failure checklist and fix map
Map	Problem Map 2.0	RAG focused recovery pipeline
Map	Problem Map 3.0	Global Debug Card, image as a debug protocol layer
Map	Semantic Clinic	Symptom to family to exact fix
Map	Grandma’s Clinic	Plain language stories mapped to Problem Map 1.0
Onboarding	Starter Village	Guided tour for newcomers
App	TXT OS	TXT semantic OS, fast boot
App	Blah Blah Blah	Abstract and paradox Q and A built on TXT OS
App	Blur Blur Blur	Text to image with semantic control
App	Blow Blow Blow	Reasoning game engine and memory demo

If this repository helped, starring it improves discovery so more builders can find the docs and tools.