Family of four open MoE language models (19B to 280B), reportedly pruned from DeepSeek and Qwen bases.
Updated EU status checked Last shown
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.
Groot is a family of four open mixture-of-experts language models published by a group called GrootTeam on Hugging Face. According to a single YouTube creator summary, sizes range from 19B total parameters (3B active) to 280B total (16B active). The models are said to be pruned and fine-tuned from DeepSeek 4.1 and Qwen 3.6 35B.
The same source claims the smallest model outperforms Qwen 3.6 at about half the size, and that FP4, FP8 and GGUF quantized versions are available. These claims are unverified here, since the fetched pages contained no model card details, benchmarks, license or release date.
The models are accessed by downloading the weights from Hugging Face for local or self-hosted use. Licensing terms, modalities beyond text, and vendor identity could not be confirmed and should be checked on the model pages.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
Family of four open MoE models from 19B (3B active) to 280B (16 active), pruned and fine-tuned from DeepSeek 4.1 and Qwen 3.6 35B. The small one beats Qwen 3.6 at half the size; FP4, FP8 and GGUF versions available.
Closest neighbours by meaning, not just by category.
Model · Alibaba
Alibaba's Qwen3.8 Flash Next: a 125B-parameter MoE language model with about 6B active parameters, previewing Qwen 4 design.
Model · Z.ai
Z.ai's open-weights 753B MoE language model, a post-trained GLM-5.2 aimed at coding, agentic tasks and cybersecurity.
Model · Alibaba
Alibaba's open-weights 2.4T-parameter MoE language model (95B active) from the Qwen3.8 family, with a hosted Max version.
Model · MiniMax
Open-weights native multimodal MoE model from MiniMax, ~428B parameters (~23B active) with 1M-token context.
Model · Rednote
Open MoE language model from Rednote (Xiaohongshu) with 208B total and 16B active parameters.
Model
Open model family (9B dense, 35B MoE, 397B MoE) trained with a self-improvement loop for reasoning, coding and agentic tasks.