Data as of October 13, 2026 · How the score is built
ScaleDown Compress
Decision readingScaleDown Compress is tracked, but not publicly ranked yet. The profile exposes 0 sourced benchmark rows and leaves unsupported fields blank until a published record exists.
ScaleDown Compress will be repriced, updated or retired. Get each notice with its source and date. Follow model changes
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Base entry
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- Not publishedScaleDown API pricing and FAQ
- Context window
- Not publishedScaleDown model limits and pricing FAQ
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- textScaleDown Compress API documentation
- Output modalities
- textScaleDown Compress API documentation
- Parameters
- Not disclosed by the provider
- Availability
- ScaleDown API · Enterprise licensingScaleDown model limits and pricing FAQ
- Cloud regions
- Not tracked yet
- Lifecycle
- Tracked
- API capabilities
- POST /compress/raw/ returns compressed context separately from the original prompt. Compression rate accepts auto or a numeric fraction of tokens to remove.ScaleDown Compress API documentation
- Prompt caching
- Not documented in the pricing recordScaleDown API pricing and FAQ
- Self-host
- Weights are not published
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
We track ScaleDown Compress, but no weighted text-model benchmark result is published on the site yet. This page shows the metadata and separate protocol evidence we can verify now; a BenchLM score will appear only if compatible public evaluations land.
ScaleDown Compress is a proprietary model. No explicit reasoning mode is documented in this profile.
Available through the ScaleDown API. Enterprise licensing offers deployment on customer hardware; public downloadable weights are not listed.
ScaleDown Compress reduces supporting context before a downstream model call through POST /compress/raw/. The response returns compressed context and the original prompt separately. The model_used field identifies the tokenizer used to count tokens, not the compression model. The provider describes the service as accepting unlimited context; no finite architectural context limit or request model ID is published in the reviewed documentation. Parameter count and exact launch date are not published in the reviewed sources. Task-specific provider results are not mapped into general benchmark scores.
The profile has no source-displayable benchmark row yet.
Last updated October 13, 2026. Runtime fields remain blank until a sourced snapshot exists.
Questions
How does ScaleDown Compress perform overall in AI benchmarks?
ScaleDown Compress does not have any source-displayable benchmark rows yet, so this profile does not assign a public score or rank. Documented specifications remain visible, while score-led charts and claims stay unavailable until a published evaluation can be attached to the exact model.
Does ScaleDown Compress have full benchmark coverage on BenchLM?
No. ScaleDown Compress currently has 0 source-displayable rows across 670 tracked benchmark slots. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of ScaleDown Compress?
ScaleDown Compress's provider has not published a context-window limit for this exact model. The profile leaves the value unavailable instead of borrowing a number from an earlier family member or a third-party route. Maximum output length is tracked separately when the provider documents it.