Maxwell
Maxwell's 35B Sparse Model Brings Long-Context AI Cost Control
Maxwell open-sourced Maxwell-V1-Flash, a 35B MoE activating 3B parameters with a 256K-token context window to cut long-context inference costs.
1 article tagged with this topic
Maxwell open-sourced Maxwell-V1-Flash, a 35B MoE activating 3B parameters with a 256K-token context window to cut long-context inference costs.