Skip to content
latestai.eulatestai.eu
G2

Gemma 2 2B

by Google

Hugging Face

Google's 2B-parameter Gemma 2 instruction-tuned model, repackaged as a text encoder for PixelDiT and PiD workflows in ComfyUI.

Available in EULanguage ModelsModel

Updated EU status checked Last shown

Gemma 2 2B at a glance

What is Gemma 2 2B?
Gemma 2 2B is a model from Google. Google's 2B-parameter Gemma 2 instruction-tuned model, repackaged as a text encoder for PixelDiT and PiD workflows in ComfyUI.
Who makes Gemma 2 2B?
Gemma 2 2B is made by Google.
Does Gemma 2 2B work in the EU?
Available in the EU. Weights are freely downloadable from Hugging Face with no regional restriction or EU exclusion in the license text shown. Last verified 14 Nov 2026.
How much does Gemma 2 2B cost?
Open source. Free to download from Hugging Face; run locally in ComfyUI. Licence: Gemma Terms of Use (original); repackage listed as nsclv1 / other.
Where was Gemma 2 2B shown?
In 1 creator video by AI Search. Most recently by AI Search on 2 Jun 2026, at 6:51 in “The BEST AI for 4K images. Free & fast”.

Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.

About

Gemma 2 2B is a small open-weights language model from Google, released in 2024 as part of the Gemma 2 family. The entry here refers to the repackaged instruction-tuned variant (gemma_2_2b_it_elm) distributed by Comfy-Org in the PixelDiT repository, where it serves as the text encoder for the PixelDiT text-to-image model and the NVIDIA PiD upscaling/latent-decode workflows.

The files are provided in two precisions: bf16 (about 5.2 GB) and fp8 scaled (about 2.6 GB). They are placed in ComfyUI's models/text_encoders folder and loaded alongside the PixelDiT or PiD diffusion model files. In this context the model encodes prompts rather than generating text.

The Comfy-Org repackaging is listed under the license name "nsclv1" (license: other), tied to the NVIDIA base models; the original Gemma 2 weights are governed by Google's Gemma terms of use. Check both licenses before commercial use. The model is intended for users running local ComfyUI image workflows.

Key features
  • Text encoder for PixelDiT and PiD ComfyUI workflows
  • About 2B parameters, instruction-tuned Gemma 2 variant
  • bf16 (5.2 GB) and fp8 scaled (2.6 GB) files
  • Single-file safetensors for ComfyUI text_encoders folder
1 moment

As seen on YouTube

Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.

Keep exploring

Similar AI tools

Closest neighbours by meaning, not just by category.

G3

Gemma 3 12B

Model · Google

1

Google's open-weights 12B multimodal (text and image input) instruction-tuned model from the Gemma 3 family.

Available in EULanguage Models
12 Mar 2025Open source
G4

Gemma 4 12B

Model · Google

2

Google's open-weights 12B encoder-free multimodal model in the Gemma 4 family, taking text, image and audio input.

Available in EULanguage Models
Release date n/aOpen source
GE

Gemma

Model · Google

2

Google DeepMind's family of open-weights models built from Gemini technology, from phone-sized to laptop-class variants.

Available in EULanguage Models
Feb 2024Open source
EG

Embedding Gemma 2

Model · Google

1

Google's small open embedding model that maps text, code, images, video and audio into a shared embedding space.

Available in EUPlatforms & Infra
Release date n/aOpen source
M2

MiniCPM5 2B

Model · OpenBMB

1

Dense 2B-parameter open-weights language model from OpenBMB for on-device and edge deployment, Apache-2.0 licensed.

Available in EULanguage Models
6 Sept 2026Open source
M2

Marlin 2B

Model · NemoStation

1

2B-parameter open-weights video-language model that writes scene descriptions and timestamped event captions.

Available in EULanguage Models
13 May 2026Open source