Cognition Releases Flash Model Post-Trained on Inkling-small
Why we picked it
The base-model choice is the interesting detail: Cognition building on Inkling-small rather than a frontier lab's model signals that smaller post-trained bases can suffice for coding-agent workloads. Builders evaluating cost-efficient models for agent pipelines should watch for Cognition's follow-up benchmarks before switching. No pricing or availability details were given in the announcement.
Cognition has released a new flash model. Soumith Chintala disclosed on X that the model was post-trained on top of Inkling-small. The tweet did not include benchmark results or parameter details.
Original · Soumith Chintala (PyTorch)
Cognition's new flash model is post-trained on top of Inkling-small!
