ED

vllm-d3-edge

EdgeGGUF by ggml-orgUpdated 2026-10-19
Download on Hugging Face
Parameters748M
Context256K
Layers12
Architectureqwen35
Smallest quant0.8 GB
Quick answer

vllm-d3-edge needs about 1.5 GB of memory at Q8_0 (a 0.8 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 0.8 GB (Q8_0) to 1.5 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 0.8 GB.

vllm-d3-edge: 0.748B multimodal decision model

vllm-d3-edge is a 0.748B multimodal foundation model based on qwen35. It answers multiple-choice, yes/no, and scoring questions in a single call, returning probabilities for each option. The model supports text, images, and videos as input.

Good at

  • Multimodal decision tasks
  • Single-call multi-question answering
  • Fast inference on GPU

Limits

  • No text generation
  • Requires 8k context for memory estimates
  • License is apache-2.0

Best for

  • Decision support systems
  • Rapid multi-choice QA
  • Embedded vision-enabled inference

Who it’s for. Developers building decision engines that need quick, probability‑based answers from text, images, or videos. The model is ideal for applications that require a single forward pass for multiple question types.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q8_00.8 GB 1.5 GB
Fits Get file →
BF161.5 GB 2.2 GB
Fits Get file →

For 8 GB VRAM, use the Q8_0 quant (0.81 GB file, ~1.5 GB memory at 8k context). If more memory is available, BF16 (1.51 GB file, ~2.2 GB memory) offers higher precision. Memory estimates are approximate.

What GPU do you need for vllm-d3-edge?

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: the best quant that fits is BF16 (1.5 GB file, about 2.2 GB in use).

Which vllm-d3-edge quant should you choose?

Q8_0 (0.8 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters748M
Layers12
Trained context262,144 tokens
KV cache (fp16)24 KB per token

Source

Quantized byggml-org
FamilyEdge
Base modelvllm-sr/d3-edge
Licenseapache-2.0
Downloads (HF)3,559
Last updated2026-10-19

Frequently asked questions about vllm-d3-edge

How much VRAM does vllm-d3-edge need?

About 1.5 GB at Q8_0 with an 8K context window. That is the 0.8 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 vllm-d3-edge on an 8 GB GPU?

Yes, with BF16: the 1.5 GB file needs about 2.2 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of vllm-d3-edge GGUF?

It depends on the quant: Q8_0 is 0.8 GB, BF16 is 1.5 GB.

What context length does vllm-d3-edge support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.2 GB at 8K and 0.8 GB at 32K.

Where can I download vllm-d3-edge GGUF?

The files are on Hugging Face in the ggml-org/vllm-d3-edge-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 vllm-d3-edge use?

Hugging Face lists the license as apache-2.0. 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 vllm-d3-edge on 8 GB VRAM?

Yes, the Q8_0 quant uses about 1.5 GB of VRAM at 8k context, fitting within an 8 GB GPU.

Does vllm-d3-edge generate text?

No, it returns probabilities for each answer option and does not generate text.

What input formats does it support?

It accepts text, images (PNG, JPEG, WebP) and videos (MP4, WebM, MOV, MKV) in various forms such as paths, URLs, base64, or arrays.

Similar models

Browse all →

More from ggml-org

All by ggml-org →

More Edge models

All Edge →

Other tiny models

All tiny models →