mancitrus/system-prompts-and-models-of-ai-tools
1
0
Fork 0
mirror of https://github.com/x1xhlol/system-prompts-and-models-of-ai-tools.git synced 2026-06-18 23:39:34 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
e34cc729aa
system-prompts-and-models-o.../salesflow-saas/memory/providers/routing-strategy.md
Claude a329957a3b
feat: Add AI engine, PDPL compliance, sequences, CPQ, and governance layers 
Phase 1-6 implementation for Dealix AI Revenue OS:

- AI Arabic Engine: NLP (arabic_nlp.py), lead scoring (lead_scoring.py)
- PDPL Compliance: consent manager, data rights handler, consent model
- Sequence Engine: multi-channel sequences with WhatsApp/Email/SMS
- CPQ System: quote engine, AI proposal generator
- Security Gate: pre-release checks, PDPL message validation
- Tool Verification: agent action audit trail
- Project Operating Files: AGENTS.md, CLAUDE.md
- Project Memory: architecture, ADRs, provider routing, PDPL checklist
- Design System: IBM Plex Sans Arabic tokens, RTL-safe components
- Sequence/Consent models for database

https://claude.ai/code/session_01LsnvBa7HwF5hs99VZbgLGj
2026-04-11 07:40:39 +00:00

1.4 KiB
Raw Blame History

LLM Provider Routing Strategy

Type: provider-config Date: 2026-04-11 Status: active

Provider Stack

Provider Use Case Latency Cost Arabic Quality
Groq (llama-3.1-70b) Fast classification, scoring ~200ms Free tier Good
Groq (llama-3.1-8b) Simple tasks, routing ~100ms Free tier Adequate
OpenAI (gpt-4o-mini) Fallback, complex reasoning ~1-2s $0.15/1M in Very Good
OpenAI (gpt-4o) Premium tasks, proposals ~2-3s $2.50/1M in Excellent
Claude (via API) Sales copy, proposals ~2-3s $3/1M in Excellent
DeepSeek Code generation ~1-2s Low N/A

Routing Rules

  1. Intent Detection: Groq llama-3.1-8b (speed priority)
  2. Lead Scoring: Groq llama-3.1-70b (accuracy needed)
  3. Arabic NLP: Groq llama-3.1-70b (good Arabic, fast)
  4. Message Writing: OpenAI gpt-4o-mini (quality Arabic output)
  5. Proposal Generation: Claude (best long-form Arabic)
  6. Conversation Summary: Groq llama-3.1-70b (speed + quality balance)
  7. Forecasting: OpenAI gpt-4o-mini (reasoning needed)

Fallback Chain

Primary → Secondary → Emergency:

  • Groq → OpenAI gpt-4o-mini → local cached response
  • OpenAI → Groq → error with retry

Cost Budget

  • Target: < $50/month for 100 active tenants
  • Groq free tier covers ~80% of requests
  • OpenAI handles remaining 20% premium tasks