9dbd7e0ba9
Overhaul of the backend AI module to produce better recipes, better images, more reliably, and cheaper. New src/ai/ module: - prompts.ts: long 'Chef Antoine' system prompt (~1500 tokens) with explicit originality rules, technical precision requirements, vocal transcription handling, and 3 few-shot style examples. Long enough to benefit from OpenAI's automatic prompt caching (-50% on cached portion from the 2nd call onward). - recipe-generator.ts: uses Structured Outputs (json_schema strict). Rich schema: titre, description, origine_inspiration, ingredients with quantity/notes/complement flag, numbered etapes with per-step duration, conseils array, accord_boisson. No more JSON.parse crashes. - image-generator.ts: switched from dall-e-3 to gpt-image-1 (medium quality by default). Much better photographic realism. Dedicated magazine-style prompt (editorial food photography, 45-deg overhead, natural light, stoneware). Slugify preserves extended Latin chars (cote-de-boeuf not c-te-de-b-uf). - transcriber.ts: migrated from whisper-1 to gpt-4o-mini-transcribe (50% cheaper, better on French). Includes a context prompt to bias toward culinary vocabulary. - cost.ts: centralized pricing table + helpers. Every OpenAI call now emits a structured log with model, durationMs, costUsd, usage, and cacheHit flag. Plugin refactor: - plugins/ai.ts now delegates to src/ai/* and only keeps the Fastify decoration glue + storage fallback for audio. - OpenAI client configured with maxRetries=3, timeout=60s. - Image generation runs in parallel with the recipe flatten/serialize step (minor speedup, ~0.5s). - flattenRecipe() converts the rich structured recipe into the legacy flat RecipeData shape (for Prisma columns) while preserving the structured form in recipeData.structured. Routes: - recipes.ts stores the structured JSON in generatedRecipe (instead of the aplatissement lossy), enabling future frontends to render rich recipes with per-ingredient notes and step timers. Env vars: - OPENAI_TRANSCRIBE_MODEL, OPENAI_IMAGE_MODEL, OPENAI_IMAGE_QUALITY, OPENAI_IMAGE_SIZE, OPENAI_MAX_RETRIES, OPENAI_TIMEOUT_MS Cost per recipe (estimated): - Before: ~$0.044 (whisper $0.003 + 4o-mini $0.0004 + dall-e-3 $0.04) - After : ~$0.018 (4o-mini-transcribe $0.0015 + 4o-mini $0.0004 + gpt-image-1 medium $0.0165), ~-59%. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
50 lines
1.3 KiB
Plaintext
50 lines
1.3 KiB
Plaintext
# ---- Requis ----
|
|
DATABASE_URL="file:./prisma/dev.db"
|
|
JWT_SECRET="change-me-please-use-at-least-32-characters"
|
|
OPENAI_API_KEY="sk-..."
|
|
|
|
# ---- Serveur ----
|
|
PORT=3000
|
|
LOG_LEVEL=info
|
|
CORS_ORIGINS=http://localhost:5173,http://127.0.0.1:5173
|
|
FRONTEND_URL=http://localhost:5173
|
|
|
|
# ---- IA ----
|
|
# Modèle texte (recette). Recommandé : gpt-4o-mini (rapide & cheap),
|
|
# ou gpt-4o pour un cran de qualité supplémentaire.
|
|
OPENAI_TEXT_MODEL=gpt-4o-mini
|
|
|
|
# Modèle de transcription audio.
|
|
# - gpt-4o-mini-transcribe : -50% par rapport à whisper-1, meilleur en français
|
|
# - whisper-1 : ancien, à éviter sauf compat
|
|
OPENAI_TRANSCRIBE_MODEL=gpt-4o-mini-transcribe
|
|
|
|
# Génération d'image
|
|
ENABLE_IMAGE_GENERATION=true
|
|
# - gpt-image-1 (recommandé) : qualité photo nettement supérieure à dall-e-3
|
|
# - dall-e-3 : ancien, à éviter
|
|
OPENAI_IMAGE_MODEL=gpt-image-1
|
|
# Pour gpt-image-1: low | medium | high
|
|
# Pour dall-e-3: standard | hd
|
|
OPENAI_IMAGE_QUALITY=medium
|
|
OPENAI_IMAGE_SIZE=1024x1024
|
|
|
|
# Robustesse OpenAI
|
|
OPENAI_MAX_RETRIES=3
|
|
OPENAI_TIMEOUT_MS=60000
|
|
|
|
# ---- Stripe (optionnel) ----
|
|
STRIPE_SECRET_KEY=
|
|
|
|
# ---- MinIO (optionnel — fallback local sinon) ----
|
|
MINIO_ENDPOINT=
|
|
MINIO_PORT=9000
|
|
MINIO_USE_SSL=false
|
|
MINIO_ACCESS_KEY=
|
|
MINIO_SECRET_KEY=
|
|
MINIO_BUCKET=freedge
|
|
MINIO_ALLOW_SELF_SIGNED=false
|
|
|
|
# ---- Email (optionnel) ----
|
|
RESEND_API_KEY=
|