# Low-resource Optimization

UniReason-Qwen3-14B-RL is a quantized model applicable to multiple domains, especially proficient in text generation and mathematical reasoning tasks.

Schreiber-mistral-nemo-12B is a large language model fine-tuned based on mistral-nemo-kartoffel-12B, focusing on providing more powerful and accurate language processing capabilities.

CSM (Conversational Speech Model) is a 1B-parameter speech generation model developed by Sesame, capable of generating RVQ audio encoding from text and audio inputs.

Llamacpp imatrix quantized version of Dolphin-Mistral-24B-Venice-Edition, supporting multiple quantization types, suitable for text generation tasks.

This is a model based on the 🤗 transformers library, optimized using unsloth technology. Specific functions and uses require further information.

XGLM-564M is a multilingual autoregressive language model with 564 million parameters, trained on a balanced corpus of 30 languages totaling 500 billion subwords.

This model is designed to detect hallucination phenomena in language model outputs, i.e., responses that are coherent but factually incorrect or out of context.

A quantized model based on dreamgen/lucid-v1-nemo, processed with llama.cpp for various quantization levels, suitable for text generation tasks.

KunoRZN-Llama-3-3B is VinkuraAI's flagship multilingual model, supporting over 12 Indian languages and English, specifically designed for education, healthcare, and daily tasks.

HiDream-I1-Full is a text-to-image diffusion model that supports generating high-quality anime-style images from text descriptions.

A text classification model fine-tuned based on distilbert-base-uncased, achieving an F1 score of 0.6147

GigaAM-v2-CTC is a Russian automatic speech recognition (ASR) model trained using the CTC loss function and can be utilized via the Hugging Face transformers library.

This is a text ranking model used for relevance-based text sorting.

Based on the Qwen2.5-7B-Instruct foundation model, it integrates the advantages of multiple 7B-scale models through model ensemble techniques.

This is a transformers model hosted on Hugging Face Hub. Specific functions and uses are pending further details.

A speech recognition model fine-tuned on Telugu audio datasets based on OpenAI Whisper-large-v2

A text classification model fine-tuned on the IMDB dataset based on the distilbert-base-uncased model

A 107M-parameter multilingual embedding model developed by IBM, supporting 12 languages with excellent performance on the MTEB benchmark

Static quantized version based on kriton/greek-text-summarization, specialized for Greek text summarization tasks

No specific model information provided, likely a generic transformers library model

GGUF quantized version of DeepSeek-R1-Distill-Llama-70B-abliterated, suitable for local inference, offering multiple quantization options to meet different hardware requirements.

A Japanese speech recognition model fine-tuned on the Common Voice 17.0 dataset based on openai/whisper-small

A Thai speech recognition model fine-tuned based on OpenAI Whisper-medium, supporting automatic speech recognition tasks for Thai and English.

A text processing model fine-tuned from t5-small, suitable for email-related tasks

This model is a multilingual sentiment analysis model fine-tuned on an unspecified dataset based on vinai/phobert-base, supporting sentiment classification tasks for multilingual texts including Vietnamese.

Speechless is a compact open-source text-to-semantic model (1 billion parameters) designed to directly convert audio into discrete semantic tokens without relying on traditional text-to-speech (TTS) models.

A text-to-speech (TTS) model specifically designed for Egyptian Arabic, developed based on the XTTS v2 architecture

IBM Granite multilingual embedding model, supporting text embedding tasks in 12 languages, suitable for information retrieval and multilingual application scenarios.

This is a fine-tuned model based on Audio Spectrogram Transformer (AST), specifically designed for emotion classification in speech audio.

A Bengali text summarization model fine-tuned based on the MT5 model, suitable for generating concise summaries of Bengali texts.

kazRush-ru-kk is a Russian-to-Kazakh translation model based on the T5 configuration, trained on multiple open-source parallel datasets.

A multimodal query rewrite model fine-tuned based on google/paligemma-3b-pt-224

A Turkish text-to-speech model fine-tuned from Microsoft SpeechT5, capable of generating natural speech

A large language model fine-tuned on multilingual datasets based on Qwen/Qwen2.5-7B, supporting 13 language processing

This is a Korean multimodal model based on the Llama3 architecture, supporting image understanding and Korean dialogue.

A speech recognition model fine-tuned on the Common Voice 11.0 Chinese dataset based on OpenAI Whisper Tiny model

A bilingual sentence embedding model based on sentence-transformers, supporting feature extraction and sentence similarity calculation

A 5-class audio classification model fine-tuned on the superb dataset based on the wav2vec pre-trained model

An instruction-following model fine-tuned based on T5-base, specifically designed to understand and execute complex instructions

A Urdu text-to-speech model fine-tuned on Microsoft's SpeechT5 architecture, supporting Romanized input