12 ms·

The Gemini-CLI situation is poor. They did not communicate that AI Pro or AI Ultra accounts cannot be used with this API broadly earlier. I specifically remembe

by hsaliak 7mo ago

The Gemini-CLI situation is poor. They did not communicate that AI Pro or AI Ultra accounts cannot be used with this API broadly earlier. I specifically remember searching for this info. Seeing this made me wonder if I had missed it. Turns out it was added to the TOS 2 days ago - diff https://github.com/google-gemini/gemini-cli/pull/20488/changes https://github.com/google-gemini/gemini-cli/pull/20488/chang.... I'd be happy to stand corrected here. Anti Gravity I understand, they are subsidizing to promote a general IDE, but I dont understand constraining the generative AI backend that Gemini CLI hits. Finally, it's unclear what's allowed and what's not if I purchase the API access from google cloud here https://developers.google.com/gemini-code-assist/docs/overview#supported-features https://developers.google.com/gemini-code-assist/docs/overvi... The Apache License of this product at this point is rich. Just make it closed source and close the API reference. Why have it out there?

tempest_ 7mo ago
I have a Code Assist Standard license to evaluate gemini-cli (and the new models) To this day I cannot coax the gemini-cli to allow me to use the models they claim you have access to. Enabled all the preview stuff in cloud etc etc. Still I mostly get 2.5 and rarely get 3 or 3.1 offered. The gemini-cli repo is a shit show. I can seem to access the new models using opencode, but am 429 rate limited almost immediately such that its like 5 minutes between calls.
- hsaliak 7mo ago
  It takes your query, computes the complexity of the request, and tries to route it to the appropriate model. There is a /manual command i think, to pick the right model. They mask the 429s well in Gemini-Cli - if an endpoint is rate limited, they try another, or route to another model, etc to keep service availability up. Your experience on the 429s is consistent with mine - the 429s is the first thing they need to fix. Fix that and they have a solid model at a good price point. I use my own coding agent (https://github.com/hsaliak/std_slop https://github.com/hsaliak/std_slop) and not being able to bring my (now cancelled) AI account with Google to it is a bummer. I'd still use it with the Code Assist Standard license if the google cloud API subscription allows for it but I have no clarification.
  - tempest_ 7mo ago
    > It takes your query, computes the complexity of the request, and tries to route it to the appropriate model. There is a /manual command i think, to pick the right model. That is what is should do, but there is no > 2.5 model shown in /model and it always picks a 2.5 model. Ive enabled preview models in the google cloud project as well. If I pass the 3 model in start param it shows 3 in the lower right corner but it is still using 2.5. I know google has issues dealing with paying customers but the current state is a shit show. If you go to the gemini-cli repo its a deluge of issues and ai slop. It seems there is a cadre of people jumping to be the first person to pump an issue into claude and get some sort of PR clout. It might be good but it needs more time to cook, or they need to take a step back and evaluate what they should consider a paid product.

by hsaliak 7mo ago

tempest_ 7mo ago

I have a Code Assist Standard license to evaluate gemini-cli (and the new models) To this day I cannot coax the gemini-cli to allow me to use the models they claim you have access to. Enabled all the preview stuff in cloud etc etc. Still I mostly get 2.5 and rarely get 3 or 3.1 offered. The gemini-cli repo is a shit show. I can seem to access the new models using opencode, but am 429 rate limited almost immediately such that its like 5 minutes between calls.

hsaliak 7mo ago
It takes your query, computes the complexity of the request, and tries to route it to the appropriate model. There is a /manual command i think, to pick the right model. They mask the 429s well in Gemini-Cli - if an endpoint is rate limited, they try another, or route to another model, etc to keep service availability up. Your experience on the 429s is consistent with mine - the 429s is the first thing they need to fix. Fix that and they have a solid model at a good price point. I use my own coding agent (https://github.com/hsaliak/std_slop https://github.com/hsaliak/std_slop) and not being able to bring my (now cancelled) AI account with Google to it is a bummer. I'd still use it with the Code Assist Standard license if the google cloud API subscription allows for it but I have no clarification.
- tempest_ 7mo ago
  > It takes your query, computes the complexity of the request, and tries to route it to the appropriate model. There is a /manual command i think, to pick the right model. That is what is should do, but there is no > 2.5 model shown in /model and it always picks a 2.5 model. Ive enabled preview models in the google cloud project as well. If I pass the 3 model in start param it shows 3 in the lower right corner but it is still using 2.5. I know google has issues dealing with paying customers but the current state is a shit show. If you go to the gemini-cli repo its a deluge of issues and ai slop. It seems there is a cadre of people jumping to be the first person to pump an issue into claude and get some sort of PR clout. It might be good but it needs more time to cook, or they need to take a step back and evaluate what they should consider a paid product.