Hebrew AI Models: A Comprehensive Catalog

A comprehensive catalog of AI models for the Hebrew language, covering LLMs, speech-to-text, text-to-speech, NLP, embeddings, translation, and more. This is a community resource maintained on GitHub.

GitHub: danielrosehill/Hebrew-AI-Models

GitHub Repository danielrosehill/Hebrew-AI-Models A catalog of AI models for the Hebrew language — LLMs, STT, TTS, NLP, embeddings, translation, and more ★ 29Updated Apr 2026 View on GitHub

Large Language Models (LLMs)

dicta-il (Israel Center for Text Analysis)

DictaLM-3.0-24B — Latest generation Hebrew LLM, 24B parameters with Base, Instruct, and Thinking variants
DictaLM-3.0-Nemotron-12B — Nemotron-based Hebrew LLM with Base and Instruct variants
DictaLM-3.0-1.7B — Compact Hebrew LLM with Base, Instruct, and Thinking variants
DictaLM 2.0 — Previous-gen Hebrew LLM (Mistral-based), 7B with Instruct variant
BEREL 3.0 — Biblical/Rabbinic Hebrew language model

Hugging Face Model dicta-il/DictaLM-3.0-24B-Base <img src="https://i.ibb.co/5Lbwyr1/dicta-logo.jpg" width="300px"/> ↓ 2.1k♥ 9Text generationAPACHE-2.0Updated May 2026 View model on Hugging Face Hugging Face Model dicta-il/DictaLM-3.0-Nemotron-12B-Base <img src="https://i.ibb.co/5Lbwyr1/dicta-logo.jpg" width="300px"/> ↓ 189Text generationUpdated Dec 2025 View model on Hugging Face Hugging Face Model dicta-il/DictaLM-3.0-1.7B-Base <img src="https://i.ibb.co/5Lbwyr1/dicta-logo.jpg" width="300px"/> ↓ 572♥ 1Text generationAPACHE-2.0Updated Dec 2025 View model on Hugging Face Hugging Face Model dicta-il/dictalm2.0 <img src="https://i.ibb.co/5Lbwyr1/dicta-logo.jpg" width="300px"/> ↓ 757♥ 28Text generationAPACHE-2.0Updated Jul 2024 View model on Hugging Face Hugging Face Model dicta-il/BEREL_3.0 Avi Shmidman, Joshua Guedalia, Shaltiel Shmidman, Cheyn Shmuel Shmidman, Eli Handel, Moshe Koppel, "Introducing BEREL: BERT Embeddings for Rabbinic-Encoded Language", Aug 2022 [arXiv:2208.01875] ↓ 4.7k♥ 5Text generationAPACHE-2.0Updated Jun 2025 View model on Hugging Face

yam-peleg

Hebrew-Gemma-11B-V2 — Hebrew-adapted Gemma 11B with V2, Instruct, and Base variants
Hebrew-Mistral-7B — Hebrew-adapted Mistral 7B, including 200K extended context version

Hugging Face Model yam-peleg/Hebrew-Gemma-11B-V2 An updated version of Hebrew-Gemma-11B that was trained longer and had some bugs fixes. ↓ 173♥ 13Text generationUpdated Mar 2024 View model on Hugging Face Hugging Face Model yam-peleg/Hebrew-Mistral-7B Hebrew-Mistral-7B is an open-source Large Language Model (LLM) pretrained in hebrew and english pretrained with 7B billion parameters, based on Mistral-7B-v1.0 from Mistral. ↓ 465♥ 73Text generationAPACHE-2.0Updated Apr 2024 View model on Hugging Face

Other Hebrew LLMs

Hebrew_Nemo — Hebrew fine-tune of Mistral Nemo 12B
HebrewGPT (Slasky) — Hebrew-native GPT trained from scratch in 1B and 296M sizes
hebrew-math-tutor-v1 (Intel) — Hebrew math tutoring model, 4B
Llama-3.2-3B-Hebrew-Master — Hebrew fine-tune of Llama 3.2

Hugging Face Model SicariusSicariiStuff/Hebrew_Nemo HebrewNemo is a state-of-the-art (SOTA) Hebrew language large language model specifically optimized for Hebrew language understanding and generation. Built upon the Mistral Nemo architecture, this model represents a… ↓ 47♥ 3Text generationAPACHE-2.0Updated Dec 2025 View model on Hugging Face Hugging Face Model Slasky/HebrewGPT-1B HebrewGPT-1B is a 1.08 billion parameter autoregressive language model trained from scratch on 2.48 billion tokens of Hebrew text. It is the first open-source, Hebrew-native GPT model of this scale, featuring a custom… ↓ 54Text generationAPACHE-2.0Updated Apr 2026 View model on Hugging Face Hugging Face Model Intel/hebrew-math-tutor-v1 Hebrew Math Tutor is a specialized mathematical reasoning model that provides step-by-step solutions to math problems in Hebrew. Built on Qwen3-4B-Thinking-2507, this model bridges the gap between advanced AI… ↓ 43♥ 9Text generationAPACHE-2.0Updated Jan 2026 View model on Hugging Face Hugging Face Model yonioz123/Llama-3.2-3B-Hebrew-Master This llama model was trained 2x faster with Unsloth and Huggingface's TRL library. ↓ 30Text generationAPACHE-2.0Updated Jan 2026 View model on Hugging Face

Speech-to-Text / ASR

ivrit-ai

The leading Hebrew ASR project, community-driven. GitHub: ivrit-ai/ivrit.ai

GitHub Repository ivrit-ai/ivrit.ai View on GitHub

whisper-large-v3-turbo-ct2 — Top Hebrew ASR model, CTranslate2 optimized (22K+ downloads)
whisper-large-v3 — Hebrew fine-tuned Whisper large-v3
pyannote-speaker-diarization-3.1 — Hebrew speaker diarization

Hugging Face Model ivrit-ai/whisper-large-v3-turbo-ct2 This is ivrit.ai's faster-whisper model, based on the ivrit-ai/whisper-large-v3-turbo Whisper model. ↓ 20.4k♥ 28Automatic speech recognitionAPACHE-2.0Updated Oct 2025 View model on Hugging Face Hugging Face Model ivrit-ai/whisper-large-v3 This model is a Hebrew finetune (continued training) of the OpenAI Whisper Large v3 model. ↓ 6.8k♥ 27Automatic speech recognitionAPACHE-2.0Updated May 2025 View model on Hugging Face Hugging Face Model ivrit-ai/pyannote-speaker-diarization-3.1 This model is a fork of pyannote/speaker-diarization-3.1. Read the original model page for all details, including citation format. ↓ 4.2k♥ 5Automatic speech recognitionMITUpdated Aug 2025 View model on Hugging Face

imvladikon

wav2vec2-xls-r-300m-hebrew — Most downloaded Hebrew model (144K downloads)
wav2vec2-xls-r-1b-hebrew — Hebrew fine-tuned XLS-R 1B

Hugging Face Model imvladikon/wav2vec2-xls-r-300m-hebrew This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the private datasets in 2 stages - firstly was fine-tuned on a small dataset with good samples Then the obtained model was fine-tuned on a large… ↓ 1.2M♥ 7Automatic speech recognitionUpdated Sept 2023 View model on Hugging Face Hugging Face Model imvladikon/wav2vec2-xls-r-1b-hebrew This model is a fine-tuned version of facebook/wav2vec2-xls-r-1b on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.3533 - Wer: 0.2251 ↓ 745♥ 2Automatic speech recognitionAPACHE-2.0Updated Sept 2023 View model on Hugging Face

Text-to-Speech (TTS)

An emerging area for Hebrew with active research:

HebTTS (Technion) — Diacritic-free Hebrew TTS research
israwave — Hebrew TTS project
Zonos-Hebrew — Hebrew variant of Zonos TTS
hebrewTTS-orpheus3b — Hebrew Orpheus TTS 3B
phonikud — Hebrew diacritization/phonemization for TTS pipelines
mms-tts-heb — Hebrew MMS TTS

GitHub Repository slp-rl/HebTTS View on GitHub GitHub Repository thewh1teagle/israwave View on GitHub Hugging Face Model notmax123/Zonos-Hebrew Zonos-v0.1 is a leading open-weight text-to-speech model trained on more than 200k hours of varied multilingual speech, delivering expressiveness and quality on par with—or even surpassing—top TTS providers. ↓ 300♥ 3Text to speechCC-BY-NC-4.0Updated Sept 2025 View model on Hugging Face Hugging Face Model Trojanssafsdg/hebrewTTS-orpheus3b This llama model was trained 2x faster with Unsloth and Huggingface's TRL library. ♥ 1APACHE-2.0Updated Mar 2025 View model on Hugging Face Hugging Face Model thewh1teagle/phonikud ↓ 64♥ 1CCUpdated Aug 2025 View model on Hugging Face Hugging Face Model thewh1teagle/mms-tts-heb View model on Hugging Face

NLP Models

BERT-based Foundation Models

AlephBERT-base (onlplab/BIU) — Foundational Hebrew BERT (29.6K downloads)
heBERT — Hebrew BERT with NER and sentiment analysis variants
DictaBERT — Comprehensive Hebrew NLP toolkit with NER, sentiment, morphology, parsing, diacritization
HeRo — Hebrew RoBERTa with long-context variant

Hugging Face Model onlplab/alephbert-base State-of-the-art language model for Hebrew. Based on Google's BERT architecture (Devlin et al. 2018). ↓ 25.9k♥ 22Fill maskAPACHE-2.0Updated Jun 2022 View model on Hugging Face Hugging Face Model avichr/heBERT from transformers import pipeline fillmask = pipeline( "fill-mask", model="avichr/heBERT", tokenizer="avichr/heBERT" ) fillmask("הקורונה לקחה את [MASK] ולנו לא נשאר דבר.") ↓ 3.2k♥ 17Fill maskUpdated Apr 2022 View model on Hugging Face Hugging Face Model avichr/heBERT_NER HeBERT is a Hebrew pretrained language model. It is based on Google's BERT architecture and it is BERT-Base config. <br> ↓ 6k♥ 5Token classificationUpdated Jan 2022 View model on Hugging Face Hugging Face Model avichr/heBERT_sentiment_analysis HeBert was trained on three datasets: 1. A Hebrew version of OSCAR (Ortiz, 2019): ~9.8 GB of data, including 1 billion words and over 20.8 million sentences. 2. A Hebrew dump of Wikipedia: ~650 MB of data, including… ↓ 27.9k♥ 26Text classificationUpdated Dec 2021 View model on Hugging Face Hugging Face Model dicta-il/dictabert State-of-the-art language model for Hebrew, released here. ↓ 7.7k♥ 12Fill maskCC-BY-4.0Updated Dec 2023 View model on Hugging Face Hugging Face Model HeNLP/HeRo State-of-the-art RoBERTa language model for Hebrew. ↓ 69♥ 6Fill maskUpdated May 2023 View model on Hugging Face Hugging Face Model HeNLP/LongHeRo State-of-the-art Longformer language model for Hebrew. ↓ 11♥ 2Fill maskUpdated May 2023 View model on Hugging Face

Embeddings and Translation

neodictabert-bilingual-embed — Bilingual Hebrew-English embeddings
sentence-transformers-alephbert — Hebrew sentence embeddings (4.9K downloads)
opus-mt-en-he (Helsinki-NLP) — English to Hebrew translation
opus-mt-tc-big-he-en (Helsinki-NLP) — Hebrew to English, large model

Hugging Face Model dicta-il/neodictabert-bilingual-embed This is a sentence-transformers model finetuned from dicta-il/neodictabert-bilingual on the he dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual… ↓ 2.1k♥ 2Sentence similarityCC-BY-4.0Updated Feb 2026 View model on Hugging Face Hugging Face Model imvladikon/sentence-transformers-alephbert This is a sentence-transformers model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. ↓ 192.6k♥ 9Sentence similarityUpdated Apr 2023 View model on Hugging Face Hugging Face Model Helsinki-NLP/opus-mt-en-he ↓ 3.1k♥ 6TranslationAPACHE-2.0Updated Aug 2023 View model on Hugging Face Hugging Face Model Helsinki-NLP/opus-mt-tc-big-he-en Neural machine translation model for translating from Hebrew (he) to English (en). ↓ 4.2k♥ 6TranslationCC-BY-4.0Updated Oct 2023 View model on Hugging Face

Key Organizations

dicta-il — Israel Center for Text Analysis. Most comprehensive Hebrew NLP toolkit
ivrit-ai — Leading Hebrew speech recognition, community-driven
onlplab (Bar-Ilan University) — AlephBERT foundational model
slprl (Technion) — Speech Processing Lab, TTS research
imvladikon — wav2vec2 Hebrew (144K downloads), sentence transformers
Helsinki-NLP — OPUS-MT Hebrew translation models

The full catalog includes OCR/vision models, speech foundation models, summarization, curated Hugging Face collections, and more. Visit the GitHub repository for the complete list.