mirror of
https://github.com/x1xhlol/system-prompts-and-models-of-ai-tools.git
synced 2026-06-18 23:39:34 +00:00
a329957a3b
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
37 lines
1.4 KiB
Markdown
37 lines
1.4 KiB
Markdown
# 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
|