Cross Layer Topk CLT is an open-source embedding model created by ByteDance. It is a cross-layer transcoder designed to read from and write to the residual stream of a base language model.
Input
Output
Context
-
Max Output
n/a
Parameters
-
Input Modalities
Output Modalities
None the catalog can show from stated facts or measured scores.
Public resources
Specification sources
As of Nov 17, 2026.
Answered from the facts on this page, with their sources.
ByteDance has not published a context window for Cross Layer Topk CLT. The catalog does not estimate one.
Yes. Cross Layer Topk CLT is an open-weight model under the Apache 2.0 license. You can download and self-host it within the license terms.
ByteDance has not published a parameter count for Cross Layer Topk CLT. The catalog does not estimate undisclosed sizes.
Strengths: Open weights under a permissive license (apache-2.0). Each is computed from the model's stated facts and benchmark results.
Cross Layer Topk CLT accepts Text and produces Embedding. The Capabilities card lists the API features it supports, such as function calling and structured output.
No benchmark source has published a result for Cross Layer Topk CLT yet, so it is shown as unmeasured, not estimated.
Not on Inferbase Managed today. Add your own provider key or endpoint that serves Cross Layer Topk CLT, and Inferbase runs it on your account with the same routing and logs.
ByteDance has not published a parameter count for Cross Layer Topk CLT, so the catalog cannot compute its memory. The GPU planner can size it from a count you enter.
Start on the free plan with 5,000 requests a month, no card required.