// Add, update, or remove text/image/video models. Handles any provider.
[HINT] Download the complete skill directory including SKILL.md and all related files
| name | model-management |
| description | Add, update, or remove text/image/video models. Handles any provider. |
.env and secrets/env.json (sops) with credentials⚠️ Pricing depends on BOTH model AND provider. Always verify pricing on the provider's website.
| File | Purpose |
|---|---|
gen.pollinations.ai/secrets/{dev,staging,prod}.vars.json | Encrypted text+image provider keys (use sops) |
gen.pollinations.ai/src/text/configs/modelConfigs.ts | Model routing config |
gen.pollinations.ai/src/text/availableModels.ts | Service name → config mapping |
shared/registry/text.ts | Pricing, provider, aliases, description |
| File | Purpose |
|---|---|
image.pollinations.ai/.env | API keys, endpoints |
gen.pollinations.ai/secrets/prod.vars.json | Encrypted secrets (use sops) |
image.pollinations.ai/src/createAndReturnImages.ts | Model handlers |
shared/registry/image.ts | Pricing, provider, aliases, description |
| Action | .env | Config/Handler | Registry |
|---|---|---|---|
| New model | ✅ | ✅ | ✅ (pricing!) |
| Change endpoint only | ✅ | - | - |
| Change provider | ✅ | ✅ | ✅ (pricing!) |
| Make paid-only | - | - | ✅ (paidOnly: true) |
| Disable model | - | ✅ (remove) | ✅ (remove) |
| Upgrade model | ✅ (if provider changes) | ✅ | ✅ (pricing!) |
After updating model files, run these tests from enter.pollinations.ai/:
# 1. Alias resolution (fast, no network) — verifies aliases map to service IDs
npx vitest run test/aliases.test.ts
# 2. Integration tests for the specific model (uses VCR snapshots)
npx vitest run test/integration/text.test.ts --testNamePattern="<service-name> "
Notes:
npm run decrypt-vars first if you haven't already# Decrypt, inspect keys
sops --decrypt gen.pollinations.ai/secrets/prod.vars.json | python3 -c "import json,sys; [print(k) for k in json.load(sys.stdin)]"
# Add/update a key
sops set gen.pollinations.ai/secrets/prod.vars.json '["KEY_NAME"]' '"value"'
# Decrypt to .env (used before local dev/tests)
npm run decrypt-vars # from image.pollinations.ai/
| Resource | Region | Used for |
|---|---|---|
myceli-prod-eastus2 | East US 2 | gptimage (gpt-image-1-mini), gptimage-large (gpt-image-1.5) |
myceli-prod-swedencentral | Sweden Central | Flux Kontext, text models |
Env var: AZURE_MYCELI_PROD_EASTUS2_API_KEY (in gen.pollinations.ai/secrets/prod.vars.json)
brew install azure-cli # if not installed
az login --use-device-code
# Use thomas@myceli.ai account
Azure blocks the whole resource (not just one deployment). Signs: all gptimage calls return 403 with "temporarily blocked because we detected behavior that may violate our content policy".
Recovery steps:
az cognitiveservices model list -l <region> --query "[?model.name=='gpt-image-1-mini']" -o jsonaz cognitiveservices account create --name myceli-prod-<region> --resource-group rg-myceli-prod --kind AIServices --sku S0 --location <region>az cognitiveservices account deployment create --name <resource> --resource-group rg-myceli-prod --deployment-name gpt-image-1-mini --model-name gpt-image-1-mini --model-version 2025-10-06 --model-format OpenAI --sku-capacity 60 --sku-name GlobalStandardaz cognitiveservices account keys list --name <resource> --resource-group rg-myceli-prod --query 'key1' -o tsvsops set gen.pollinations.ai/secrets/prod.vars.json '["AZURE_MYCELI_PROD_EASTUS2_API_KEY"]' '"<key>"'createAndReturnImages.ts (AZURE_GPTIMAGE_CONFIGS)az cognitiveservices account deployment delete --name <old-resource> --resource-group rg-myceli-prod --deployment-name <deployment>npm run dev then curl -H "x-enter-token: $PLN_ENTER_TOKEN" "http://localhost:16384/prompt/a+cat?model=gptimage"