#deepseek-coder

14 messages · Page 1 of 1 (latest)

tacit lodge
#

DeepSeek coder is DeepSeek chat now, there is only one model now, not two

golden juniper
#

Cool diagram

#

I hope they're not running all these at the same time

tacit lodge
#

The leading model for coding is Sonnet-3.5 btw

graceful comet
#

Oh ok sorry I couldn't understand but now I do. 🫡

golden juniper
#

Is the hyperbolic provider 128k as well? Hyperbolic says 8k

pliant prawn
#

DeepSeek-V2.5-1210 is an upgraded version of DeepSeek-V2.5, with improvements across various capabilities:

Mathematical: Performance on the MATH-500 benchmark has improved from 74.8% to 82.8% .

Coding: Accuracy on the LiveCodebench (08.01 - 12.01) benchmark has increased from 29.2% to 34.38% .

Writing and Reasoning: Corresponding improvements have been observed in internal test datasets.

Additionally, the new version of the model has optimized the user experience for file upload and webpage summarization functionalities.

frank tide
#

Is it available through the API?

tacit lodge
frank tide
#

I mean the new one 1210 ?

tacit lodge
# frank tide I mean the new one 1210 ?

If DeepSeek (the company) updated the model on their API, they should always be the latest version. Other providers, see the (i) icon in the provider tab in the model card.