How to configure the fast mode of the GPT model?
The original poster is a native Chinese speaker. The following content has been automatically translated into English by AI.
As shown in the screenshot below, I created a new Variants and configured the gpt-5.4-high-fast model myself. I have multiple GPT Plus accounts, but I strongly prefer Claude Code over Codex, so I chose to use CCS to set up a reverse proxy for multiple GPT Plus accounts.
I want to explicitly configure the specific model variant — for example, I want to use the 5.4 model with different reasoning depths: xhigh, high, and medium. Most importantly, I need to enable the fast suffix to activate the fast mode for GPT-series models. That is the whole reason I am configuring a custom model.
The specific configuration JSON is as follows:
{
"env": {
"ANTHROPIC_MODEL": "gpt-5.4-high-fast",
"ANTHROPIC_DEFAULT_OPUS_MODEL": "gpt-5.4-xhigh-fast",
"ANTHROPIC_DEFAULT_SONNET_MODEL": "gpt-5.4-high-fast",
"ANTHROPIC_DEFAULT_HAIKU_MODEL": "gpt-5.4-medium-fast"
}
}However, an error clearly occurs:
The error message is as follows:
API Error: 502 {"error":{"message":"unknown provider for model gpt-5.4-high-fast","type":"server_error","code":"internal_server_error"}}Did I make an obvious configuration mistake somewhere? Or how should I pass the parameters to correctly configure both the GPT model variant and the fast mode?
贡献者
ruan-cat
Cursor