Search papers, labs, and topics across Lattice.
2
3
4
The DeepSeek-V4.1-Flash model, a multimodal Mixture-of-Experts model with 552B backbone parameters and support for contexts of up to one million tokens, is introduced, substantially improving cost efficiency for agentic workloads and pushing the limits of KV cache compression.
Multimodal embeddings get a serious upgrade with CoCoA, a new pre-training method that forces models to compress all input information into a single token for reconstruction, leading to substantial quality gains.
