Deca 2.5
The model built for the future.
The model built for the future.
Deca 2.5 introduces groundbreaking capabilties that redefine what's possible with AI.
Turbo models can deliver 4000 tokens/second, up to 15x faster then Deca 2.
Comparable performance with major models on all benchmarks.
Deca 2.5 is designed to be cheap, with lower cost per token than comparable major models. Use Deca 2.5 for as little as $0.02/MTok with Flex mode.
4000 t/s at zero load.
Deca 2.5 gives comparable performance to major models on all benchmarks.
Benchmarks are estimated.
Small but intelligent model for everyday tasks.
For power users who need more intelligence for less cost.
Up to 1000 tokens per second at zero load
Powerful model for tasks requiring extreme intelligence.
Normal:deca-2.5-ultra-extended)
Average prices for models with dwarf-based pricing. Prices may be higher or lower, depending on the dwarf model activated, and context.