OpenCodeGo
OpenCodeGo is a service provider that offers AI model APIs.
Timeline (last 90 days)
The author suggests that a first-month $5 subscription can continue to be used through the current month, but subscriptions at $10 or above for the following month are not recommended.
OpenCodeGo reduced all of its previously $60 subscription tiers (Flash, Pro, etc.) to $15, but with significantly reduced request quotas.
OpenCodeGo's co-founder stated on Twitter that suppliers could reproduce pre-price-increase DeepSeek pricing on rented GPUs.
Eleven days after promising not to raise prices, DeepSeek's official price increase took effect, and OpenCodeGo changed its position, claiming that suppliers had deceived them and that suppliers could no longer deliver.
Users who have subscribed to OpenCodeGo should immediately cancel the auto-renewal feature on their payment platform to avoid being charged next month.
Third-party service provider OpenCodeGo raised its prices almost simultaneously following DeepSeek's price increase.
Before DeepSeek's price increase, OpenCodeGo had publicly promised not to raise prices for the time being, stating that original price levels could be maintained through supplier channels.
After OpenCodeGo's price increase, a service previously priced at $0.43 rose to $0.66, and $0.87 rose to $1.98.
Users with graphics cards of 16GB or more can try locally deploying Qwen3 32B; users without a graphics card or who prefer convenience are recommended to use free APIs or subscribe to cheaper coding plans such as OpenCodeGo.
OpenCodeGo's DeepSeek-V4 plan quota is 7,600 requests per 5 hours and approximately 37,000 requests per month, with the first month at approximately $5 and renewals at approximately $10
2 more timeline events
All Facts (2)
Users with graphics cards of 16GB or more can try locally deploying Qwen3 32B; users without a graphics card or who prefer convenience are recommended to use free APIs or subscribe to cheaper coding plans such as OpenCodeGo.
50%UnverifiedThe OpenCodeGo commercial relay platform charges per number of requests, with model coverage ranging from Grok to the DeepSeek-V4 series
50%