Salamandra-2B-Instruct-GGUF Open-Source Model - A Powerful Tool for Text Generation Supporting Over 30 Languages

Salamandra 2b Instruct GGUF

Developed by tensorblock

A GGUF-format 2B-parameter multilingual instruction fine-tuned model supporting 30+ languages, suitable for text generation tasks.

Large Language Model

Transformers

Open Source License:Apache-2.0 #Multilingual instruction fine-tuning #Lightweight GGUF quantization #European legal text optimization

Downloads 120

Release Time : 2/15/2025

Model Overview

A 2B-parameter multilingual instruction fine-tuned model developed by BSC-LT, offering multiple quantization versions for lightweight text generation applications.

Model Features

Multilingual support

Supports 30+ European languages and code generation

Lightweight quantization

Offers 12 quantization versions from Q2_K to Q8_0 to accommodate different hardware requirements

Instruction fine-tuning

Optimized through instruction fine-tuning, suitable for dialogue and task execution

Model Capabilities

Multilingual text generation

Code generation

Instruction understanding and execution

Dialogue interaction

Use Cases

Multilingual applications

European language translation assistance

Text conversion between supported languages

Education

Programming teaching assistant

Explain code snippets and generate teaching examples

🚀 BSC-LT/salamandra-2b-instruct - GGUF

This repository offers GGUF format model files for BSC-LT/salamandra-2b-instruct, providing a practical solution for various text - generation tasks.

📄 License

The project is licensed under the Apache - 2.0 license.

🌐 Language Support

This model supports multiple languages, including Bulgarian (bg), Catalan (ca), Code, Czech (cs), Welsh (cy), Danish (da), German (de), Greek (el), English (en), Spanish (es), Estonian (et), Basque (eu), Finnish (fi), French (fr), Irish (ga), Galician (gl), Croatian (hr), Hungarian (hu), Italian (it), Lithuanian (lt), Latvian (lv), Maltese (mt), Dutch (nl), Norwegian Nynorsk (nn), Norwegian Bokmål (no), Occitan (oc), Polish (pl), Portuguese (pt), Romanian (ro), Russian (ru), Serbo - Croatian (sh), Slovak (sk), Slovenian (sl), Serbian (sr), Swedish (sv), and Ukrainian (uk).

📊 Datasets

The model is trained on the following datasets:

oscar - corpus/colossal - oscar - 1.0
HuggingFaceFW/fineweb - edu
joelniklaus/eurlex_resources
joelito/legal - mc4
projecte - aina/CATalog
UFRGS/brwac
community - datasets/hrwac
danish - foundation - models/danish - gigaword
HiTZ/euscrawl
PleIAs/French - PD - Newspapers
PleIAs/French - PD - Books
AI - team - UoA/greek_legal_code
HiTZ/latxa - corpus - v1.1
allenai/peS2o
pile - of - law/pile - of - law
PORTULAN/parlamento - pt
hoskinson - center/proof - pile
togethercomputer/RedPajama - Data - 1T
bigcode/starcoderdata
bjoernp/tagesschau - 2018 - 2023
EleutherAI/the_pile_deduplicated

⚙️ Base Model

The base model used is BSC - LT/salamandra - 2b - instruct.

🏷️ Tags

The model is tagged with TensorBlock and GGUF.

Feedback and support: TensorBlock's Twitter/X, Telegram Group and Discord server

🚀 Quick Start

This repo contains GGUF format model files for BSC - LT/salamandra - 2b - instruct. The files were quantized using machines provided by TensorBlock, and they are compatible with llama.cpp as of commit b4658.

🌟 Our Projects

Project	Description	Image	Link
Awesome MCP Servers	A comprehensive collection of Model Context Protocol (MCP) servers.		GitHub
TensorBlock Studio	A lightweight, open, and extensible multi - LLM interaction studio.		GitHub

💻 Prompt Template

<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

📋 Model File Specification

Filename	Quant type	File Size	Description
salamandra-2b-instruct-Q2_K.gguf	Q2_K	1.087 GB	smallest, significant quality loss - not recommended for most purposes
salamandra-2b-instruct-Q3_K_S.gguf	Q3_K_S	1.215 GB	very small, high quality loss
salamandra-2b-instruct-Q3_K_M.gguf	Q3_K_M	1.277 GB	very small, high quality loss
salamandra-2b-instruct-Q3_K_L.gguf	Q3_K_L	1.317 GB	small, substantial quality loss
salamandra-2b-instruct-Q4_0.gguf	Q4_0	1.410 GB	legacy; small, very high quality loss - prefer using Q3_K_M
salamandra-2b-instruct-Q4_K_S.gguf	Q4_K_S	1.447 GB	small, greater quality loss
salamandra-2b-instruct-Q4_K_M.gguf	Q4_K_M	1.506 GB	medium, balanced quality - recommended
salamandra-2b-instruct-Q5_0.gguf	Q5_0	1.626 GB	legacy; medium, balanced quality - prefer using Q4_K_M
salamandra-2b-instruct-Q5_K_S.gguf	Q5_K_S	1.642 GB	large, low quality loss - recommended
salamandra-2b-instruct-Q5_K_M.gguf	Q5_K_M	1.691 GB	large, very low quality loss - recommended
salamandra-2b-instruct-Q6_K.gguf	Q6_K	1.920 GB	very large, extremely low quality loss
salamandra-2b-instruct-Q8_0.gguf	Q8_0	2.401 GB	very large, extremely low quality loss - not recommended

📥 Downloading Instruction

Command line

Firstly, install Huggingface Client:

pip install -U "huggingface_hub[cli]"

Then, download the individual model file to a local directory:

huggingface-cli download tensorblock/salamandra-2b-instruct-GGUF --include "salamandra-2b-instruct-Q2_K.gguf" --local-dir MY_LOCAL_DIR

If you want to download multiple model files with a pattern (e.g., *Q4_K*gguf), you can try:

huggingface-cli download tensorblock/salamandra-2b-instruct-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'

Phi 2 GGUF

Other

Phi-2 is a small yet powerful language model developed by Microsoft, featuring 2.7 billion parameters, focusing on efficient inference and high-quality text generation.

Large Language Model Supports Multiple Languages

A large English language model pre-trained with masked language modeling objectives, using improved BERT training methods

Large Language Model English

FacebookAI

19.4M

212

Distilbert Base Uncased

Apache-2.0

DistilBERT is a distilled version of the BERT base model, maintaining similar performance while being more lightweight and efficient, suitable for natural language processing tasks such as sequence classification and token classification.

Large Language Model English

distilbert

11.1M

669

Llama 3.1 8B Instruct GGUF

Meta Llama 3.1 8B Instruct is a multilingual large language model optimized for multilingual dialogue use cases, excelling in common industry benchmarks.

Large Language Model English

XLM-RoBERTa is a multilingual model pretrained on 2.5TB of filtered CommonCrawl data across 100 languages, using masked language modeling as the training objective.

Large Language Model Supports Multiple Languages

An English pre-trained model based on Transformer architecture, trained on massive text through masked language modeling objectives, supporting text feature extraction and downstream task fine-tuning

Large Language Model English

OPT is an open pre-trained Transformer language model suite released by Meta AI, with parameter sizes ranging from 125 million to 175 billion, designed to match the performance of the GPT-3 series while promoting open research in large-scale language models.

Large Language Model English

facebook

6.3M

198

A pretrained model based on the transformers library, suitable for various NLP tasks

Llama 3.1 8B Instruct

Llama 3.1 is Meta's multilingual large language model series, featuring 8B, 70B, and 405B parameter scales, supporting 8 languages and code generation, with optimized multilingual dialogue scenarios.

Large Language Model

Transformers Supports Multiple Languages

The T5 Base Version is a text-to-text Transformer model developed by Google with 220 million parameters, supporting multilingual NLP tasks.

Large Language Model Supports Multiple Languages

google-t5

5.4M

702

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご