# Multilingual Instruction Following

Qwen3 is the latest generation of the Tongyi Qianwen series of large language models, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Based on large-scale training, Qwen3 has achieved breakthrough progress in reasoning, instruction following, agent capabilities, and multilingual support.

The FP8 quantized version of the Qwen3-235B-A22B model, which effectively reduces GPU memory requirements and improves computational throughput, suitable for various natural language processing scenarios.

Qwen3-14B-FP8-dynamic is an optimized large language model. By quantizing activation values and weights to the FP8 data type, it effectively reduces GPU memory requirements and improves computational throughput.

Qwen3 is the latest version of the Tongyi Qianwen series of large language models, offering a range of dense and mixture-of-experts (MoE) models with groundbreaking advancements in reasoning, instruction following, agent capabilities, and multilingual support.

Qwen3 is the latest generation of large language models in the Tongyi Qianwen series, offering a complete suite of dense models and mixture-of-experts (MoE) models. Based on large-scale training, Qwen3 achieves breakthrough progress in reasoning, instruction following, agent capabilities, and multilingual support.

Qwen3-0.6B-FP8 is the latest version in the Tongyi Qianwen series of large language models, offering a 0.6B-parameter FP8 quantized version that supports free switching of mind modes and multilingual tasks.

Granite-3.2-2B-Instruct is a 2-billion-parameter long-context AI model, fine-tuned for cognitive reasoning capabilities, supporting 12 languages and multitasking.

Granite-3.2-8B-Instruct is an 8-billion-parameter long-context AI model, specifically fine-tuned for cognitive reasoning capabilities, supporting multiple languages and tasks.

This is an orthogonalized version of microsoft/Phi-3-medium-4k-instruct, where specific techniques suppress the model's tendency to reject, preserving as much of the original model's knowledge and capabilities as possible.

Phi-3-Medium-4K-Instruct is a 14-billion-parameter lightweight open-source model focusing on high-quality reasoning capabilities, supporting 4K context length, suitable for commercial and research purposes in English environments.

Mixtral-8x7B is a pre-trained generative sparse mixture of experts large language model that outperforms Llama 2 70B on most benchmarks.

A multilingual instruction-following language model based on LLaMA 7B, supporting English, Chinese, and Japanese, suitable for chatbots and instruction-following tasks.