Government Of India
A-
A
A+
ORGANISATION sarvamtranslate
Translation model for 22 Indian Languages
1upvotes
127downloads
2,451views
0
0Shares
updated 07 Jul 2025
Upvote 1
LICENSE GNU General Public License, version 3
SECTOR Sector Agnostic
About Model
Sarvam-Translate is an advanced translation model from Sarvam AI, specifically designed for comprehensive, document-level translation across the 22 official Indian languages, built on Gemma3-4B-IT. It addresses modern translation needs by moving beyond isolated sentences to handle long-context inputs, diverse content types, and various formats. Sarvam-Translate aims to provide high-quality, contextually aware translations for Indian languages, which have traditionally lagged behind high-resource languages in LLM performance. Assamese, Bengali, Bodo, Dogri, Gujarati, English, Hindi, Kannada, Kashmiri, Konkani, Maithili, Malayalam, Manipuri, Marathi, Nepali, Odia, Punjabi, Sanskrit, Santali, Sindhi, Tamil, Telugu, Urdu
Metadata
GNU General Public License, version 3
Sector Agnostic
sarvamai
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
Open
07/08/25 06:47:53
N.A.
0
Tags
Related Models
LayoutLMv3-Finetuned-FUNSD - Form Understanding Model
A fine-tuned version of LayoutLMv3 trained on the FUNSD dataset for form understanding and key-value extraction in scanned documents.
LayoutXLM - Multilingual Document Understanding
A multimodal pre-trained Transformer model designed for multilingual document AI, integrating text, layout, and image data to improve visually rich document understanding across multiple languages.
LayoutLM-Document-QA - Visual Question Answering for Documents
A fine-tuned version of LayoutLM for document-based question answering, trained on SQuAD2.0 and DocVQA to extract answers from scanned documents, invoices, and financial reports
Table Transformer (TATR) - PubTables1M Table Structure Recognition
A Transformer-based object detection model pre-trained on PubTables1M for recognizing table structures, including row and column segmentation, in scientific and academic documents.
Table Transformer (TATR) - FinTabNet Table Structure Recognition
A Transformer-based object detection model pre-trained on FinTabNet for recognizing table structures, including row and column detection, in financial and tabular documents.
LayoutLMv2 - Multimodal Document Understanding
An improved multimodal Transformer model for document AI, integrating text, layout, and image pre-training for enhanced document understanding tasks such as form recognition, document classification, and visual question answering.
LayoutLMv3 - Multimodal Document AI
A pre-trained multimodal Transformer model for document AI, integrating unified text and image masking for tasks such as form understanding, receipt processing, and document layout analysis.
Table Transformer (TATR) - Table Structure Recognition
A Transformer-based object detection model pre-trained for table structure recognition, including row and column detection, using the PubTables1M and FinTabNet datasets.
LayoutLM - Multimodal Pre-training for Document Understanding
A multimodal Transformer model pre-trained on text and layout for document image understanding, optimized for form processing, receipt understanding, and structured document analysis.
Orca 2-13Billion Parameter Model
A 13-billion-parameter language model fine-tuned from LLaMA-2, designed to enhance reasoning capabilities in tasks such as reading comprehension, math problem-solving, and text summarization.
Related Models by Organization
shuka-v1
Multilingual audio to text model
sarvam-1
India's first indic model, pretrained on 4 trillion tokens
sarvam-30B
Sarvam-30B is an advanced Mixture-of-Experts (MoE) model with 2.4B non-embedding active parameters, designed primarily for practical deployment. It combines strong reasoning, reliable coding ability, and best-in-class conversational quality across Indian languages. Sarvam-30B is built to run reliably in resource-constrained environments and can handle multilingual voice calls while performing tool calls.
sarvam-105b
Sarvam-105B is an advanced Mixture-of-Experts (MoE) model with 10.3B active parameters, designed for superior performance across a wide range of complex tasks. It is highly optimized for complex reasoning, with particular strength in agentic tasks, mathematics, and coding.
sarvamM
Multilingual, hybrid-reasoning, text-only language model built on Mistral-Small
Related Models by Sector
LayoutLMv3-Finetuned-FUNSD - Form Understanding Model
A fine-tuned version of LayoutLMv3 trained on the FUNSD dataset for form understanding and key-value extraction in scanned documents.
LayoutXLM - Multilingual Document Understanding
A multimodal pre-trained Transformer model designed for multilingual document AI, integrating text, layout, and image data to improve visually rich document understanding across multiple languages.
LayoutLM-Document-QA - Visual Question Answering for Documents
A fine-tuned version of LayoutLM for document-based question answering, trained on SQuAD2.0 and DocVQA to extract answers from scanned documents, invoices, and financial reports
Table Transformer (TATR) - PubTables1M Table Structure Recognition
A Transformer-based object detection model pre-trained on PubTables1M for recognizing table structures, including row and column segmentation, in scientific and academic documents.
Table Transformer (TATR) - FinTabNet Table Structure Recognition
A Transformer-based object detection model pre-trained on FinTabNet for recognizing table structures, including row and column detection, in financial and tabular documents.
LayoutLMv2 - Multimodal Document Understanding
An improved multimodal Transformer model for document AI, integrating text, layout, and image pre-training for enhanced document understanding tasks such as form recognition, document classification, and visual question answering.
LayoutLMv3 - Multimodal Document AI
A pre-trained multimodal Transformer model for document AI, integrating unified text and image masking for tasks such as form understanding, receipt processing, and document layout analysis.
Table Transformer (TATR) - Table Structure Recognition
A Transformer-based object detection model pre-trained for table structure recognition, including row and column detection, using the PubTables1M and FinTabNet datasets.
LayoutLM - Multimodal Pre-training for Document Understanding
A multimodal Transformer model pre-trained on text and layout for document image understanding, optimized for form processing, receipt understanding, and structured document analysis.
Orca 2-13Billion Parameter Model
A 13-billion-parameter language model fine-tuned from LLaMA-2, designed to enhance reasoning capabilities in tasks such as reading comprehension, math problem-solving, and text summarization.
Related Models by Task Type
parrotlet e
Parrotlet-e is a state of the art multilingual medical entity embedding model designed for understanding and linking medical terms across Indian languages.
AI4Bharat - Romansetu-200M -Multilingual LLM for Indian langauges using romanization
RomanSetu is Efficiently unlocking multilingual (Indian Languages) capabilities of Large Language Models via Romanization.
AI4Bharat - Romansetu-100M - Multilingual LLM for Indian langauges using romanization
RomanSetu is Efficiently unlocking multilingual (Indian Languages) capabilities of Large Language Models via Romanization.
AI4Bharat – Romanized Path – Base to Supervised Fine-Tuning (SFT)
Romansetu model is built on base pretrained model which is supervised fine tuned on instuction-following tasks using romanized Indian languages.
AI4Bharat – Native-Script Path – Base to Supervised Fine-Tuning (SFT)
Romansetu model is built on base pretrained model which is supervised fine-tuned on instruction datasets in native scripts of Indian languages.
AI4Bharat - Romansetu-300M -Multilingual LLM for Indian Languages using Native script
RomanSetu is Efficiently unlocking multilingual capabilities of Large Language Models via Native Script.
AI4Bharat - Romansetu-200M -Multilingual LLM for Indian Languages using Native script
RomanSetu is Efficiently unlocking multilingual capabilities of Large Language Models via Native Script.
AI4Bharat - Romansetu-100M -Multilingual LLM for Indian Languages using Native script
RomanSetu is Efficiently unlocking multilingual (Indian Langauages) capabilities of Large Language Models via Native Scripts.
AI4Bharat - Romansetu-300M -Multilingual LLM for Indian langauges using romanization
RomanSetu is Efficiently unlocking multilingual capabilities (Indian Languages) of Large Language Models via Romanization.
LecaDis
Dictionary Constrained Disambiguation for Improved NMT
© 2026 - Copyright AIKosh. All rights reserved.