NE

Nemotron 3.5 Embed 1B

NemotronGGUF by maternionUpdated 2026-10-21
Download on Hugging Face
Parameters1.14B
Context256K
Layers16
Architecturemistral3
Smallest quant0.9 GB
Quick answer

Nemotron 3.5 Embed 1B needs about 2.3 GB of memory at Q8_0 (a 1.2 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 0.9 GB (Q6_K) to 2.3 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 2.1 GB.

Nemotron 3.5 Embed 1B – 1.14B parameter multilingual multimodal embedding model

Nemotron 3.5 Embed 1B is a 1.14B parameter model that produces 2048‑dimensional embeddings for text, image, or combined image‑text inputs. It supports a wide range of languages and includes a Pixtral vision projector for image embeddings. The model is available in GGUF format and can be run with llama.cpp or Ollama.

Good at

  • Multilingual text embeddings
  • Multimodal image‑text embeddings
  • Retrieval tasks

Limits

  • No vision support in Q6_K and Q8_0 without BF16 projector
  • Requires 8k context for memory estimates
  • License is OpenMDW-1.1

Best for

  • Document retrieval systems
  • Multilingual search engines
  • Embedding‑based similarity search

Who it’s for. Developers building retrieval or search applications that need multilingual and multimodal embeddings. The model is suitable for both commercial and non‑commercial use under the OpenMDW-1.1 license.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q6_K0.9 GB 2.0 GB
Fits Get file →
Q8_01.2 GB 2.3 GB
Fits Get file →
BF162.3 GB 3.3 GB
Fits Get file →

For 8 GB VRAM, use Q6_K (0.94 GB file, ~2 GB RAM). For 12 GB VRAM, Q8_0 (1.22 GB file, ~2.3 GB RAM). If higher precision is needed and 16 GB VRAM is available, BF16 (2.29 GB file, ~3.3 GB RAM) is recommended.

What GPU do you need for Nemotron 3.5 Embed 1B?

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 (2.3 GB file, about 3.3 GB in use).

Which Nemotron 3.5 Embed 1B quant should you choose?

Q8_0 (1.2 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.

If memory is tight, the smallest file is Q6_K at 0.9 GB, with the quality trade-offs that come with fewer bits. Very close to the original quality for most uses, at a clearly larger size than 4-bit.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturemistral3
Parameters1.14B
Layers16
Trained context262,144 tokens
KV cache (fp16)64 KB per token

Source

Quantized bymaternion
FamilyNemotron
Licenseother
Downloads (HF)801
Last updated2026-10-21

Frequently asked questions about Nemotron 3.5 Embed 1B

How much VRAM does Nemotron 3.5 Embed 1B need?

About 2.3 GB at Q8_0 with an 8K context window. That is the 1.2 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 Nemotron 3.5 Embed 1B on an 8 GB GPU?

Yes, with BF16: the 2.3 GB file needs about 3.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Nemotron 3.5 Embed 1B GGUF?

It depends on the quant: Q6_K is 0.9 GB, Q8_0 is 1.2 GB, BF16 is 2.3 GB.

What context length does Nemotron 3.5 Embed 1B support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.5 GB at 8K and 2.1 GB at 32K.

Where can I download Nemotron 3.5 Embed 1B GGUF?

The files are on Hugging Face in the maternion/Nemotron-3.5-Embed-1B-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 Nemotron 3.5 Embed 1B use?

Hugging Face lists the license as other. 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 Nemotron 3.5 Embed 1B on 8 GB VRAM?

Yes, the Q6_K quantization uses 0.94 GB file size and requires about 2 GB RAM at 8k context, fitting within 8 GB VRAM.

Does the model support image embeddings?

Yes, the GGUF includes a BF16 Pixtral vision projector, enabling image and document‑page embeddings.

What license applies to this model?

The model is governed by the OpenMDW License Agreement, version 1.1 (OpenMDW-1.1).

Other GGUFs of Nemotron-3.5-Embed-1B-BF16

Similar models

Browse all →

More from maternion

All by maternion →

More Nemotron models

All Nemotron →

Other tiny models

All tiny models →