VoxCPM is an innovative tokenizer-free end-to-end text-to-speech (TTS) system that overcomes the limitations of discrete tokenization by modeling speech in a continuous space. It has two core capabilities: context-aware speech generation and realistic zero-shot voice cloning. It can automatically adjust the prosody and style according to the text content and clone the speaker's timbre, accent, and emotion with just a short reference audio.

Audio ProcessingSafetensorsMultiple Languages

Qwen3-4B-Hivemind-Instruct-NEO-MAX-Imatrix-GGUF is a large language model with 4B parameters for instruction fine-tuning based on the Qwen3 architecture. It uses NEO Imatrix and MAX quantization technologies and has the ability to process an ultra-long context of 256k. This model performs excellently in multiple benchmark tests and is a powerful general-purpose model.

Natural Language ProcessingGgufEnglish

Gemma 3 is a lightweight open-source multimodal model launched by Google. This version is an instruction-tuned quantization-aware training model with 12B parameters, which has been converted to the MXFP4 format of the MLX framework. It supports text and image input and generates text output, with a 128K context window and support for over 140 languages.

MultimodalMlx

This is the GGUF quantized version of the ai-sage/GigaChat3-10B-A1.8B-bf16 model, offering a variety of quantization options, from high-precision Q8_0 to extremely compressed smol-IQ1_KT, to meet the deployment requirements under different hardware conditions. This model supports a 32K context length, adopts the MLA architecture, and is optimized for dialogue scenarios.

Natural Language ProcessingGgufOther

GigaChat3-10B-A1.8B is a dialogue model in the GigaChat series, based on the Mixture of Experts (MoE) architecture, with a total of 10 billion parameters, of which 1.8 billion are active parameters. This model uses multi-head latent attention and multi-token prediction technologies, supports a long context of 256,000 tokens, and performs excellently in multilingual dialogue and reasoning tasks.

Natural Language ProcessingSafetensorsMultiple Languages

MiroThinker v1.0 is an open-source research agent that enhances tool-enhanced reasoning and information search capabilities through model-level interactive expansion. This model performs excellently in multiple benchmark tests and supports long contexts and in-depth multi-step analysis.

Natural Language ProcessingTransformersEnglish

Dia2 is a streaming dialogue text-to-speech (TTS) model developed by Nari Labs. It supports real-time speech generation and can start generating audio without the full text. It can be adjusted according to the dialogue context to achieve a natural and smooth dialogue experience.

Audio ProcessingSafetensorsEnglish

A large language model with 12B parameters fine-tuned based on Qwen 3, capable of handling 256k ultra-long contexts. It is specifically optimized for long text generation tasks. This model combines the foundation of the VLTO 8B model, Brainstorm 20x tuning, and NEO Imatrix dataset quantization technology, enabling it to generate high-quality long text content.

Natural Language ProcessingGgufMultiple Languages

This model is a GGUF format model generated by in-depth fine-tuning of the Qwen3-Coder-30B-A3B-Instruct Mixture of Experts model using Brainstorm 20x technology. It is designed for programming and general purposes, supports a 256K long context, and is optimized for code generation and understanding of mainstream and niche programming languages.

Natural Language ProcessingGgufMultiple Languages

Tarka-Embedding-350M-V1 is a text embedding model with 350 million parameters, capable of generating 1024-dimensional dense text representations. This model is optimized for downstream applications such as semantic similarity, search, and retrieval-augmented generation (RAG), supports multiple languages, and has the ability to handle long contexts.

Natural Language ProcessingSentence-transformers

The top model in the aquif-3.5 series, with advanced inference capabilities and a large context window of 1 million tokens, and has excellent performance in multiple benchmark tests. The comprehensive AAII score reaches 60 points.

Natural Language ProcessingTransformersMultiple Languages

The aquif-3.5 series is the masterpiece released on November 3, 2025, including two versions, Plus and Max, which provide advanced inference capabilities and an unprecedented 1 million token context window, achieving state-of-the-art performance in their respective categories.

Natural Language ProcessingTransformersMultiple Languages

Olmo 3 is a brand - new 32B parameter language model family developed by the Allen Institute for AI (Ai2), which includes variants such as Base, Instruct, and Think. This model is trained on the Dolma 3 dataset, supports long - context processing of up to 65,536 tokens, and aims to promote the development of language model science. The model is fully open - source and follows the Apache 2.0 license.

Natural Language ProcessingTransformersEnglish

Kimi K2 Thinking is the latest generation of open-source thinking model developed by Moonshot AI. It has powerful deep reasoning ability and tool call function. The model adopts a mixture-of-experts architecture, supports native INT4 quantization, has a 256k context window, and performs excellently in multiple benchmark tests.

Natural Language ProcessingTransformers

Tarka-Embedding-150M-V1 is an embedding model with 150 million parameters, capable of generating 768-dimensional dense text representations. It is optimized for various downstream applications such as semantic similarity, search, and retrieval-augmented generation (RAG), focusing on capturing deep contextual semantics to support general text understanding across different domains.

Natural Language ProcessingSentence-transformers

Qwen3-VL-4B-Instruct is the latest vision-language model in the Tongyi Qianwen series. It achieves excellent visual perception, text understanding and generation, spatial reasoning, and intelligent agent interaction capabilities at the 4B parameter scale. It supports long contexts and video understanding, and has powerful OCR and multi-language processing capabilities.

MultimodalTransformers

TabPFN-2.5 is a tabular foundation model based on the Transformer architecture. It utilizes context learning technology and can solve tabular prediction problems in a single forward propagation, providing efficient regression and classification solutions for structured tabular data.

Qwen3-VL is the most powerful vision-language model in the Tongyi series. It has excellent text understanding and generation capabilities, in-depth visual perception and reasoning capabilities, long context support, strong spatial and video dynamic understanding capabilities, and outstanding agent interaction capabilities.

MultimodalGguf

Qwen3-VL-32B-Thinking is the most powerful vision-language model in the Qwen series, with excellent text understanding and generation capabilities, in-depth visual perception and reasoning capabilities, long context processing, spatial and video dynamic understanding capabilities, and outstanding agent interaction capabilities.

MultimodalTransformers

Qwen3-VL-8B-Thinking is the most powerful vision-language model in the Tongyi Qianwen series. It has excellent text understanding and generation capabilities, in-depth visual perception and reasoning capabilities, long context support, strong spatial and video dynamic understanding capabilities, and outstanding agent interaction capabilities.

MultimodalGguf