Support Request for New DeepSeek Model Implementation

·

Feature request: Support for the new DeepSeek model, which offers performance comparable to frontier models at a significantly lower cost.

11 comments

LuisArturo
·
·
Hey there Ben thanks for reaching out and thank you for the feedback, will go ahead and bring this up with the team.
🙌1
Channeled
APP
·
·
Hi Ben G.! This thread was recently closed by our Support team. If you have a moment, please share your feedback:
Nicholas J.
·
·
im also interested in this
Channeled
APP
·
·
Thank you so much for sharing your feedback Ben G.!
Mark L.
·
·
We've added new reasoning models in the "Content creation" category, including DeepSeek's R1 and OpenAi's o3-mini. https://downloads.intercomcdn.com/i/o/w28k1kwz/1365385624/15cd42f6124f4369eeda9a197303/20250204-ai-models.png?expires=1738706400&signature=2e573609eb7f5db926ddd4d94878f2e67a7b5866a2b87ffa41c6c5eb1561a580&req=dSMhE8p2mIddXfMW1HO4zUokIHi0UNHFoDH8bcaoUw5TB3vp7RVmPf%2Fu2dpu%0ANtYv%0A Enjoy!
Ben G.
·
·
What about their v3 model
Ben G.
·
·
It’s cheaper and I think people could save a lot of money with this
Mark L.
·
·
Not yet! I'm curious – could you let me know any specific use cases or workflows or challenges, where V# would be especially helpful for you? Or mostly a cost/efficiency item? I'll share feedback internally, either way, but any extra context is welcome.
Ben G.
·
·
It’s better than gpt 4o and cheaper so with a custom api key we’d actually save quite a bit
Ben G.
·
·
And adding either V3 or R1 to claygent would be even better
Mark L.
·
·
Makes total sense! And a big +1 on Claygent ability – that would be a wonderful addition. Thanks for taking the time to weigh in! Much appreciated. :)