Stephanie, If you were going to compare the top end sourcers available in Merlin, which do you think is all-around best? Best answers coupled with least idea throttling/influencing is, I suppose, how one would define best. You have used these more than most -- interested in your opinion.
As you can see, GPT 3.5, Gemini 1.5 Flash, and Mixtral are the cheapest (1x) whereas Claude Opus is the most expensive at 50x. Note that this is not a simple query to credit ratio - a longer output in Mixtral usually takes 4 credits from my usage. For a limited plan I would only use Opus when really needed. Thus, I have Mixtral set as the default. If you have an OpenAI key and AppSumo Tier 2 or up you can use your OpenAI key for GPT queries after credits are exhausted.
In terms of which is the best, I honestly don't see much difference between GPT 4 and 4o (Omni), except Omni can take visual input. GPT 4 is a little bit of a better writer than 3.5. Claude is usually better with longer text but I wouldn't use it otherwise except maybe fo fiction, because it is so heavily "moderated" and nags you about Western medicine if you try to ask about alternatives. I haven't used Gemini much but it's Google so I don't trust it.
BTW, Mixtral can apparently handle longer text (at least more so than GPT 3.5) but I haven't had a chance to test it yet.
Stephanie, If you were going to compare the top end sourcers available in Merlin, which do you think is all-around best? Best answers coupled with least idea throttling/influencing is, I suppose, how one would define best. You have used these more than most -- interested in your opinion.
Here is the "query standards" data they use to charge credits - https://www.getmerlin.in/query-standards
As you can see, GPT 3.5, Gemini 1.5 Flash, and Mixtral are the cheapest (1x) whereas Claude Opus is the most expensive at 50x. Note that this is not a simple query to credit ratio - a longer output in Mixtral usually takes 4 credits from my usage. For a limited plan I would only use Opus when really needed. Thus, I have Mixtral set as the default. If you have an OpenAI key and AppSumo Tier 2 or up you can use your OpenAI key for GPT queries after credits are exhausted.
In terms of which is the best, I honestly don't see much difference between GPT 4 and 4o (Omni), except Omni can take visual input. GPT 4 is a little bit of a better writer than 3.5. Claude is usually better with longer text but I wouldn't use it otherwise except maybe fo fiction, because it is so heavily "moderated" and nags you about Western medicine if you try to ask about alternatives. I haven't used Gemini much but it's Google so I don't trust it.
BTW, Mixtral can apparently handle longer text (at least more so than GPT 3.5) but I haven't had a chance to test it yet.