LLM EXPLORER 60,828 MODELS INDEXED

Recurrentgemma 2B It by google

By google · 4026 downloads
HF Repo ↗ Copy URL

Recurrentgemma 2B It is an open-source language model by google. Features: 2b LLM, VRAM: 5.4GB, License: gemma, LLM Explorer Score: 0.2, Arc: 31, HellaSwag: 56.3, MMLU: 40.9, GSM8K: 10.1.

  Arxiv:1705.03551   Arxiv:1804.06876   Arxiv:1804.09301   Arxiv:1809.02789   Arxiv:1811.00937   Arxiv:1904.09728   Arxiv:1905.07830   Arxiv:1905.10044   Arxiv:1907.10641   Arxiv:1911.01547   Arxiv:1911.11641   Arxiv:2009.03300   Arxiv:2009.11462   Arxiv:2101.11718   Arxiv:2103.03874   Arxiv:2107.03374   Arxiv:2108.07732   Arxiv:2109.07958   Arxiv:2110.08193   Arxiv:2110.14168   Arxiv:2203.09509   Arxiv:2206.04615   Arxiv:2304.06364   Arxiv:2402.19427   Conversational   Endpoints compatible   Recurrent gemma   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/google/recurrentgemma-2b-it 

Recurrentgemma 2B It Parameters and Internals

Model Type 
open language model, text generation, instruction-tuned
Use Cases 
Areas:
Content creation and communication, Research and education
Applications:
Text generation, Chatbots and conversational AI, Text summarization, NLP research, Language Learning Tools, Knowledge Exploration
Primary Use Cases:
Creative text generation, Conversational interfaces, Text summarization
Limitations:
Bias due to training data, Complex task difficulty, Ambiguities in language, Factual inaccuracies, Lacking common sense
Considerations:
Biases, task complexity, factual accuracy, and responsible use were considered.
Supported Languages 
English (Proficient)
Training Details 
Methodology:
Recurrent architecture
Hardware Used:
TPUv5e
Model Architecture:
Recurrent architecture
Safety Evaluation 
Methodologies:
Red-teaming, Human evaluation
Findings:
Safe within acceptable thresholds
Risk Categories:
Text-to-text content safety, Text-to-text representational harms, Memorization, Large-scale harm
Ethical Considerations:
Bias, misinformation, misuse, transparency, and accountability were considered.
Responsible Ai Considerations 
Fairness:
The model underwent careful scrutiny and is reported in the model card.
Transparency:
Details on architectures, capabilities, and limitations are shared.
Accountability:
Developers should execute responsibilities with internal policies and guidelines.
Mitigation Strategies:
De-biasing techniques, user guidelines, and privacy-preserving techniques were explored.
Input Output 
Input Format:
Text string
Accepted Modalities:
Text
Output Format:
Generated text
LLM NameRecurrentgemma 2B It
Repository 🤗https://huggingface.co/google/recurrentgemma-2b-it 
Model Size2b
Required VRAM5.4 GB
Updated2026-07-23
Maintainergoogle
Model Typerecurrent_gemma
Model Files  5.0 GB: 1-of-2   0.4 GB: 2-of-2
Model ArchitectureRecurrentGemmaForCausalLM
Licensegemma
Transformers Version4.40.0.dev0
Tokenizer ClassGemmaTokenizer
Padding Token<pad>
Vocabulary Size256000
Torch Data Typebfloat16

Best Alternatives to Recurrentgemma 2B It

Best Alternatives
Context / RAM
Downloads
Likes
Recurrentgemma 2B0K / 10.8 GB212898
...ntgemma 2B It Bnb 4bit Smashed0K / 2.5 GB81