# 128K long text processing

Phi-3.5-mini-instruct is a lightweight and advanced open-source model built on the dataset used by Phi-3, focusing on high-quality, inference-rich data. It supports a 128K token context length and has powerful multilingual and long-context processing capabilities.

Typhoon2.1-Gemma3-12B is an instruction-based Thai large language model with 12 billion parameters, a context length of 128K, and function call ability.

Typhoon2.1-Gemma3-4B is an instruction-based large Thai language model with 4 billion parameters and a context length of 128K, supporting function call ability.

Qwen2.5 is the latest series of Tongyi large language models, offering a variety of models with parameters ranging from 500 million to 72 billion, with significant improvements in coding, math, instruction following, and long text processing.

Qwen3 is the latest generation of large language models in the Qwen series, offering a range of dense and mixture-of-experts (MoE) models. Based on extensive training, Qwen3 has made breakthroughs in reasoning, instruction following, agent capabilities, and multilingual support.

Plain text adaptation of Gemma 3 27B, removing image processing capabilities to focus on text generation tasks

Qwen2.5 is the latest series of the Tongyi Qianwen large model, including base models and instruction-tuned models with parameter scales ranging from 0.5B to 72B, showing significant improvements in areas such as code, mathematics, instruction following, and long text generation.

Gemma 3 is a lightweight multimodal model launched by Google, capable of processing text and image inputs and generating text outputs. This model has a 128K large context window and multilingual support for over 140 languages.

Gemma is a lightweight open model series launched by Google, built on the same research technology as Gemini. The llamafile version is packaged as an executable file by Mozilla for easy use on multiple platforms.

FP4 quantized version of DeepSeek R1 model, using optimized Transformer architecture for efficient text generation

Command R+ 08-2024 is a 104 billion parameter multilingual large language model that supports Retrieval-Augmented Generation (RAG) and tool usage, suitable for automating complex tasks.

Cohere Labs Command R+ is a multilingual large language model with 104 billion parameters, featuring advanced capabilities such as retrieval-augmented generation (RAG) and tool usage, and supporting a context length of 128K.