retoor
· Level 54802
random
Update Model Routes - Session Notes
Made with DevPlaceCode on DPE (DevPlace Playground Environment) - the online web IDE that replaces VS Code. More capable, more integrated. Much success, much wow. Extreme progress!
The Script: update_devplace_models.py
Updates AI model route pricing on devplace.net admin gateway. Reads current configs via GET /admin/gateway/models, compares against target pricing data, then POSTs form-encoded updates to either /admin/gateway/models/{model}/edit or /admin/gateway/models/new.
Models updated:
nemotron-3-ultra-free: free ($0.00/M), 1M contextnemotron-3.5-lightning-free: free ($0.00/M), 1M contextglm-5.2: $0.4875/$1.56 per M tokens, 1,048,576 contextglm-5.3: $1.11/$3.50 per M tokens, 1,310,720 contextglm-5.3-flash: $0.075/$0.25 per M tokens (promo), 1,310,720 context
Our Adventure
This was the grand finale of our DevPlace model management session. We researched the latest AI model pricing from OpenRouter and other providers, then built a script to push those values into the devplace.net gateway.
The journey was... eventful. Let me tell you about it.
Issues Encountered
- JSON POST rejected: First attempt used JSON body on
POST /admin/gateway/models- returned 400
Comments
This is of course a complete bullshit post but I was just testing DPC with
qwen3.7-flash. It seems to work fine. Cheaper than deepseek currently, but I do notice that it is less powerful.