StartLux Decision v1 4B BF16 needs about 10.0 GB of memory at BF16 (a 8.4 GB file) with an 8K context window, so it fits on a 12 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
StartLux Decision v1 4B BF16 – a 4.21B parameter decision model
StartLux Decision v1 4B BF16 is a 4.21B parameter model built on the qwen35 architecture. It answers typed questions about a state, returning probabilities for each option. The model supports up to 262,144 tokens and includes a vision tower for image evidence.
Good at
- Typed decision making
- High token context
- Vision‑enabled inputs
Limits
- No commercial license
- No explicit vision support in weights
- Requires recent llama.cpp
Best for
- Decision‑making APIs
- Batch question answering
- Vision‑augmented state analysis
Who it’s for. Researchers and developers needing a non‑commercial decision model with large context and image support. The CC BY‑NC 4.0 license allows free use in research and non‑commercial projects.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| BF16 | 8.4 GB | 10.0 GB | Too big | Get file → |
The only available quantization is BF16, 8.42 GB in size. It requires about 10 GB of VRAM at 8 k context. For lower VRAM budgets, no other quant is provided, so the BF16 file is the only option.
What GPU do you need for StartLux Decision v1 4B BF16?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: the best quant that fits is BF16 (8.4 GB file, about 10.0 GB in use).
Which StartLux Decision v1 4B BF16 quant should you choose?
BF16 (8.4 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about StartLux Decision v1 4B BF16
How much VRAM does StartLux Decision v1 4B BF16 need?
About 10.0 GB at BF16 with an 8K context window. That is the 8.4 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run StartLux Decision v1 4B BF16 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 10.0 GB, so you would need a larger GPU or CPU offloading.
What is the file size of StartLux Decision v1 4B BF16 GGUF?
It depends on the quant: BF16 is 8.4 GB.
What context length does StartLux Decision v1 4B BF16 support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.
Where can I download StartLux Decision v1 4B BF16 GGUF?
The files are on Hugging Face in the startlux-models/StartLux-Decision-v1-4B-BF16-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does StartLux Decision v1 4B BF16 use?
Hugging Face lists the license as cc-by-nc-4.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run StartLux Decision v1 4B BF16 on 8 GB VRAM?
No. The BF16 file needs about 10 GB of VRAM at 8 k context, so 8 GB is insufficient.
Does the model support image inputs?
Yes. The weights include a vision tower and the inference package accepts images as part of the evidence.
Is the model suitable for commercial use?
No. The CC BY‑NC 4.0 license allows only non‑commercial use; commercial use requires a separate license.