Last Updated: 2026-02-02

Status: ✅ Complete

File: lib/env.ts

• Update default POLLINATIONS_BASE_URL → https://gen.pollinations.ai
• Add POLLINATIONS_CHAT_PATH (default: /v1/chat/completions)
• Remove POLLINATIONS_MODEL env var (now dynamic)
• Add optional POLLINATIONS_PRIMARY_MODEL for override

Status: ✅ Complete

File: lib/db.ts

• Add model_cache table for discovered models
• Add model_config table for selected primary/fallback models
• Add upsertModelCache(model) function
• Add getModelCache(type) function
• Add getModelConfig(key) / setModelConfig(key, value) functions
• Add getModelByName(name) function
• Add clearModelCache(type?) function
• Add getAllModelConfig() function

Status: ✅ Complete

File: lib/llm_pollinations.ts

• Add discoverTextModels() - GET /v1/models + /text/models
• Add discoverImageModels() - GET /image/models
• Add refreshModelCache() - run discovery and cache to SQLite
• Add selectBestModel(options) - pick model based on needs
• Add getModelWithFallbacks() - return primary + fallback list
• Add isDiscoveryNeeded() - check if cache is stale (>24h)
• Add ensureModelCache() - run discovery if needed

Status: ✅ Complete

File: lib/llm_pollinations.ts

• Update generateChat() with model hints (useTools, complexTask)
• Add auto-retry with fallback models on error
• Add modelUsed and fallbacksAttempted to response
• Update generateJSON() to use new wrapper
• Update generateText() to use new wrapper
• Handle content_blocks in Gemini responses

Status: ✅ Complete

File: lib/llm_pollinations.ts

• Add generateImage(prompt, options) function
• Download image and store in Val Town blob storage
• Add getImageModels() function

File: app/http.ts

• Add POST /api/image/generate endpoint
• Add GET /api/image/models endpoint

Status: ✅ Complete

File: jobs/sync/interval.ts

• Add model cache refresh (daily)
• Check last_discovery timestamp before refreshing
• Log model discovery results

Status: ✅ Complete

File: lib/agent.ts

• Use complexTask: true for plan generation
• Log which model was used
• Use cheaper model for simple agent interactions

Status: ✅ Complete

File: app/http.ts

• Add GET /api/ai/models - list cached models + config
• Add POST /api/ai/discover - trigger model rediscovery

• GET /api/ai/models - List cached text/image models and current config
• POST /api/ai/discover - Manually trigger model rediscovery
• POST /api/image/generate - Generate image from prompt (stored in blob)
• GET /api/image/models - List available image models

• model_cache - Stores discovered models with metadata
• model_config - Stores selected primary/fallback models

1. Primary: gemini (cheap, tools=true)
2. Fallback 1: openai (baseline compatibility)
3. Fallback 2: gemini-large (complex tasks)

• Model discovery runs automatically once per day via interval job
• Models are auto-selected based on task complexity
• Automatic fallback on model errors
• Images stored in Val Town blob storage

# Required
READWISE_TOKEN=your_readwise_token
APP_ADMIN_SECRET=your_secret_for_api_auth
POLLINATIONS_TOKEN=your_pollinations_sk_token

# Optional (with defaults)
POLLINATIONS_BASE_URL=https://gen.pollinations.ai
POLLINATIONS_CHAT_PATH=/v1/chat/completions
POLLINATIONS_PRIMARY_MODEL=  # leave empty for auto-discovery

# Optional (webhooks)
READWISE_WEBHOOK_SECRET=