LE

lev

LevGGUF by ggml-orgUpdated 2026-10-19
Download on Hugging Face
Parameters4.2B
Context256K
Layers32
Architectureqwen35
Smallest quant3.0 GB
Quick answer

lev needs about 4.6 GB of memory at Q4_K_M (a 3.0 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 3 quants, from 3.0 GB (Q4_K_M) to 8.4 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

Decision model for high‑volume judgement calls

lev is a Qwen3.5‑4B LoRA adapter that answers typed questions in a single forward pass. It returns calibrated probabilities over a supplied option set, with no generated tokens. The model is licensed under Apache‑2.0 and supports the TypeSafe `/v1/systemone` protocol.

Good at

  • Zero‑token output for fast decisions
  • Calibrated probabilities for gating
  • Single‑pass multi‑question handling

Limits

  • Requires CUDA GPU for real‑time use
  • No vision or text generation capabilities
  • Large base model (~9.3 GB) must be downloaded

Best for

  • Routing and moderation workflows
  • Intent detection and triage
  • Grading and LLM output checking

Who it’s for. Developers building high‑throughput decision systems that need fast, calibrated answers without generating text. The model is ideal for product teams integrating TypeSafe SDKs and requiring deterministic option selection.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_M3.0 GB 4.6 GB
Fits Get file →
Q8_04.5 GB 6.1 GB
Fits Get file →
BF168.4 GB 10.0 GB
Too big Get file →

For 8 GB VRAM, use Q4_K_M (3.03 GB, ~4.6 GB memory at 8k context). For 12 GB VRAM, Q8_0 (4.48 GB, ~6.1 GB memory). BF16 is only suitable for larger memory budgets.

What GPU do you need for lev?

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: the best quant that fits is Q4_K_M (3.0 GB file, about 4.6 GB in use).
  • 8 GB VRAM: the best quant that fits is Q8_0 (4.5 GB file, about 6.1 GB in use).
  • 12 GB VRAM: the best quant that fits is BF16 (8.4 GB file, about 10.0 GB in use).

Which lev quant should you choose?

Q4_K_M is the usual starting point for lev: it balances file size and quality well. Here it is a 3.0 GB file.

The Q8_0 file is 4.5 GB, about 1.5 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters4.2B
Layers32
Trained context262,144 tokens
KV cache (fp16)128 KB per token

Source

Quantized byggml-org
FamilyLev
Licenseapache-2.0
Downloads (HF)8,311
Last updated2026-10-19

Frequently asked questions about lev

How much VRAM does lev need?

About 4.6 GB at Q4_K_M with an 8K context window. That is the 3.0 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 lev on an 8 GB GPU?

Yes, with Q8_0: the 4.5 GB file needs about 6.1 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of lev GGUF?

It depends on the quant: Q4_K_M is 3.0 GB, Q8_0 is 4.5 GB, BF16 is 8.4 GB.

What context length does lev 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 lev GGUF?

The files are on Hugging Face in the ggml-org/lev-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 lev 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 lev on 8 GB VRAM?

Yes, using the Q4_K_M quantization which requires about 4.6 GB of VRAM at 8k context.

Does lev generate text?

No, it returns probabilities over supplied options and produces no output tokens.

Similar models

Browse all →

More from ggml-org

All by ggml-org →

More Lev models

All Lev →

Other small models

All small models →