Uyu 2 28B is an open-source language model by mente-ai. Features: 31b LLM, VRAM: 56.4GB, Context: 256K, License: apache-2.0, LLM Explorer Score: 0.38.
Base model:finetune:google/gem... Base model:google/gemma-4-31b-... Conversational Custom code En English Gemma4 text Gisp Ko Korean Pruning Region:us Roleplay Safetensors Sharded Tensorflow Vllm
Model Card on HF 🤗: https://huggingface.co/mente-ai/uyu-2-28B
Uyu 2 28B Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
Uyu 2 28B Parameters and Internals
| LLM Name | Uyu 2 28B |
| Repository 🤗 | https://huggingface.co/mente-ai/uyu-2-28B |
| Base Model(s) | |
| Model Size | 31b |
| Required VRAM | 56.4 GB |
| Updated | 2026-07-12 |
| Maintainer | mente-ai |
| Model Type | gemma4_text |
| Model Files | |
| Supported Languages | ko en |
| Model Architecture | Uyu2ForCausalLM |
| License | apache-2.0 |
| Context Length | 262144 |
| Model Max Length | 262144 |
| Transformers Version | 5.12.1 |
| Tokenizer Class | GemmaTokenizer |
| Padding Token | <pad> |
| Vocabulary Size | 262144 |
Rank the Uyu 2 28B Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 55029 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a