gpt-oss-safeguard-120b is a security reasoning model built on gpt-oss, specifically designed for security use cases. This model can classify text content based on the provided security policies and perform a series of basic security tasks. It is suitable for scenarios such as input-output filtering of large language models and online content annotation.

Natural Language ProcessingTransformers

The GPT-OSS series is an open-weight model launched by OpenAI, designed for powerful inference capabilities, agent tasks, and diverse developer use cases. This model can effectively solve complex problems, provide a flexible and efficient usage experience, and support the Apache 2.0 open-source license.

Natural Language ProcessingTransformers

The GPT-OSS series is an open-weight model released by OpenAI, designed for powerful reasoning, agent tasks, and diverse developer use cases. The 120B version has 117 billion parameters (5.1 billion of which are active parameters), suitable for production environments and scenarios requiring high reasoning ability.

Natural Language ProcessingTransformers

Qwen3 is the latest version in the Tongyi Qianwen series of large language models, offering a 1.7B-parameter GPTQ 8-bit quantized model that supports switching between reasoning and non-reasoning modes, enhancing inference capabilities and multilingual support.

Natural Language ProcessingTransformers

GLM-4-32B-0414 is a new member of the GLM family with a parameter scale of 32 billion, comparable in performance to the GPT series, and supports local deployment.

Natural Language ProcessingTransformersMultiple Languages

GLM-4-32B-0414 is a new member of the GLM family, featuring 32 billion parameters with performance comparable to the GPT series and DeepSeek-V3 series, supporting local deployment.

Natural Language ProcessingTransformersMultiple Languages

GLM-4-32B-0414 is a new member of the GLM family, featuring 32 billion parameters, with performance comparable to the GPT and DeepSeek series, and supports local deployment.

Natural Language ProcessingTransformersMultiple Languages

GLM-4-32B-0414 is a large language model with 32 billion parameters, comparable in performance to the GPT series, supporting both Chinese and English, and excels in code generation, function calling, and complex task processing.

Natural Language ProcessingTransformersMultiple Languages

MiniCPM3-4B is the third-generation model in the MiniCPM series, with overall performance surpassing Phi-3.5-mini-Instruct and GPT-3.5-Turbo-0125, comparable to several recent 7B~9B-scale models.

Natural Language ProcessingTransformersMultiple Languages

TEMPO is a time series forecasting model based on the GPT-2 architecture, achieving efficient prediction through prompt engineering.

Natural Language ProcessingTransformersEnglish

PAGnol is a series of large-scale French language models for free text generation, with 1.5 billion parameters. Based on the GPT architecture, it was developed in collaboration between LightOn and Inria's ALMAnaCH team.

Natural Language ProcessingTransformersFrench

OPT is an open pre-trained Transformer language model developed by Meta AI, containing 1.3B parameters, comparable to the GPT-3 series models

Natural Language ProcessingTransformersEnglish

Aira-2 is the second version of the Aira instruction tuning series, a Portuguese instruction-tuned model based on GPT-2, suitable for text generation and dialogue tasks.

Natural Language ProcessingTransformersportuguese

LaMini-GPT-1.5B is a large language model fine-tuned based on the GPT-2-xl architecture, belonging to the LaMini-LM series, focusing on instruction-following tasks

Natural Language ProcessingTransformersEnglish

LaMini-GPT-124M is a member of the LaMini-LM model series. Based on the GPT-2 architecture, it is fine-tuned on a dataset of 2.58 million instructions and is specifically designed for natural language instruction response. This model has 124M parameters and can understand and respond to human natural language instructions, providing users with answers to relevant questions.

Natural Language ProcessingTransformersEnglish

A 111M parameter model in the Cerebras-GPT series, adopting GPT-3 style architecture, trained on The Pile dataset, achieving compute-optimal performance following Chinchilla scaling laws.

Natural Language ProcessingTransformersEnglish

JASMINE is a series of Arabic GPT models designed for few-shot learning, with parameters ranging from 300 million to 6.7 billion, pretrained on 235GB of text data.

Natural Language ProcessingTransformers

OPT is an open-source large language model series launched by Meta AI, benchmarked against the GPT-3 architecture, aimed at promoting reproducibility and societal impact discussions in large model research.

Natural Language ProcessingJaxEnglish

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.

Natural Language ProcessingJaxEnglish