Skip to main content
Radar

Keep up with the models you depend on. Follow price changes, retirements, and API updates.Follow the models you depend on.

Follow model changes
BenchLM

BTL-4 Compact

CurrentReleased Aug 5, 2026Open WeightReasoning262K contextBad Theory Labs BTL-4 Compact model card

Released Aug 5, 2026 — see all recent releases

Decision reading
BTL-4 Compact is tracked, but not publicly ranked yet. The profile exposes 0 sourced benchmark rows and leaves unsupported fields blank until a published record exists.

Data as of September 10, 2026 · How the score is built

Compare BTL-4 CompactFind alternatives

Share or export

Share on XLinkedInSocial cardBenchmarks CSVJSON

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
badtheorylabs/BTL-4-CompactBad Theory Labs BTL-4 Compact model card
Context window
262KBad Theory Labs BTL-4 Compact model card
Maximum output
Not sourced yet
Knowledge cutoff
Not sourced yet
Input modalities
Not sourced yet
Output modalities
Not sourced yet
Parameters
Not sourced yet
Availability
A single 9.96 GB IQ2_XXS GGUF (2.30 bits per weight) of the 35.1B-total, 2.1B-active BTL-4 mixture of experts, published under Apache 2.0 for llama.cpp with the full 262,144-token context.
Cloud regions
Not tracked yet
Lifecycle
Current
API capabilities
Tool calling, structured outputs, and batch support are not tracked yet
Prompt caching
Not documented in the pricing recordBad Theory Labs BTL-4 Compact model card
Self-host
Open weights available; hardware estimate not sourced
Rate limits
Not tracked yet

Deployment options

Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.

Published weights are available, but BenchLM does not yet have a sourced parameter and VRAM profile for this exact model. Hardware cost estimates stay unavailable until that sizing record is complete.

Estimate VRAM from known parameters

Lineage

The sequence follows explicit supersedes links. A successor's displayed score stays at least 0.1 points above its predecessor; raw benchmark rows do not move. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.

  1. Aug 5, 2026 · you are here

    BTL-4 Compact

    Not publicly ranked · Price not listed

Compact

BTL-4BTL-4BTL-3

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 BTL-4 Compact, 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.

BTL-4 Compact is a open weight model with a 262K context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.

A single 9.96 GB IQ2_XXS GGUF (2.30 bits per weight) of the 35.1B-total, 2.1B-active BTL-4 mixture of experts, published under Apache 2.0 for llama.cpp with the full 262,144-token context.

Bad Theory Labs’ August 6, 2026 card reports that the Compact build retains 94.1% of the full-precision BTL-4’s measured behaviour but publishes no per-benchmark scores for it, so BenchLM stores no scored rows; see the BTL-4 row for the full-precision results. The row stays unranked.

BTL-4 Compact sits in the BTL family with BTL-4, BTL-3. The profile has no source-displayable benchmark row yet.

Frequently asked questions

How does BTL-4 Compact perform overall in AI benchmarks?

BTL-4 Compact 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.

Is BTL-4 Compact open source?

BTL-4 Compact is an open-weight model from Bad Theory Labs. Its weights can be downloaded for local or hosted deployment, subject to the published license. Open weight does not automatically mean open source: training data and training code may remain private, and commercial restrictions can still apply.

Which sibling models are related to BTL-4 Compact?

BTL-4 Compact belongs to the BTL family. Related tracked variants include BTL-4, BTL-3. A sibling link indicates shared lineage or a documented configuration relationship; it does not mean the variants have identical pricing, context limits, benchmark evidence, or deployment behavior. Compare before switching.

Does BTL-4 Compact have full benchmark coverage on BenchLM?

No. BTL-4 Compact currently has 0 source-displayable rows across 434 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 BTL-4 Compact?

BTL-4 Compact has a documented context window of 262K. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.

Compare BTL-4 Compact with every tracked model482 comparisons

Last updated September 10, 2026. Runtime fields remain blank until a sourced snapshot exists.

Watch BTL-4 Compact in the weekly brief

Get one weekly email when material rank, price, availability, or benchmark evidence changes are worth revisiting.

Read a sample issue

Join 2,000+ readers.