Typing Mind

Currently, when we go to models and set the temperature to zero, it affects all models. If I want to use a model like OpenAI o1, it only supports temperature 1, so it throws an error. I have to set it to 1 for all models, or else one by one, adjust each chat.

I’d like to know what model does currently Gemini 1.5 Pro 2M points to (I mean, for example gemini-1.5-pro or gemini-1.5-pro-latest). I’d also like to request to add the two new Gemini models to typingmind (Gemini 1.5 Pro 002 and Gemini 1.5 Flash 002).

https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching This would enable massive savings when dealing with long conversations.

Would be awesome to have Gemini Pro and Gemini Pro Vision models to be supported via API in TypingMind: https://cloud.google.com/vertex-ai/docs/generative-ai/model-reference/gemini#gemini-pro

Claude recently started to show <thinking> </thinking> . Since we likely pay for this output when it's not essential, is it possible to not show/use this part of the tokens used to reduce cost?

https://x.com/alexalbert__/status/1812921642143900036 Sonnet 3.5 can have apparently double the output it currently has allowed in your tool

Add cost estimation for gemini models especially 1.5 pro and flash

Provide access to the models hosted at AWS Bedrock using IAM profiles and keys. Since Anthropic doesn't give out many Claude API keys, this is a viable alternative to people wanting Claude without having to share an API key with many other users.

https://platform.openai.com/docs/models/gpt-4o