A large MoE model with 456B parameters, a massive context window of 1,000,000 tokens, and a reasoning budget of 40,000 tokens. Thanks to architectural innovations, the model is more resource-efficient compared to models of similar size, making it highly effective for a wide range of intelligent analysis tasks and agent-based applications.
DeepSeek-R1-0528-Qwen3-8B is a compact model based on Qwen3 with 8 billion parameters, distilled from the flagship version DeepSeek-R1-0528. It achieves state-of-the-art (SOTA) results among open-source models in its category. The model is ideally suited for deployment in resource-constrained environments while retaining advanced mathematical and logical reasoning capabilities from the teacher model.
DeepSeek-R1-0528 is the first major update to the popular DeepSeek R1 series, released on May 28, 2025. The developers revised their approach to depth of thought, and the number of parameters increased to 685 billion, resulting in an improvement of more than 10 percentage points across nearly all significant benchmarks compared to the version released on January 22, 2025.
VisualClozePipeline-384 is one of the models compatible with diffusion models within the VisualCloze framework for image generation with visual context.
Phi-4-Reasoning is a compact 14-billion-parameter reasoning model that confidently competes with much larger models in mathematics, programming, and scientific tasks. The model is ideally suited for educational and research applications where high-quality logical reasoning is required while efficiently utilizing computational resources.
Qwen3-32B — the flagship dense model with 32 billion parameters and a context window of 128K tokens, designed for mission-critical AI systems. It delivers state-of-the-art quality in the most complex tasks and is ideal for building advanced AI products.
Qwen3-14B is a model with 14 billion parameters and a context window of 128K tokens, delivering performance comparable to flagship solutions. It is ideally suited for tasks requiring expert-level analysis and content generation with heightened attention to detail.
Qwen3-8B is the most frequently downloaded model in the Qwen3 series on Hugging Face. It supports switching between thinking modes and delivers the best performance in its scale, significantly surpassing Qwen2.5-7B in overall capabilities.
Qwen3-4B is a compact 4-billion-parameter model featuring an extended 32K token context window. Remarkably, developers claim its performance rivals that of the much larger Qwen2.5-72B-Instruct. This model is particularly well-suited for analytical tasks, technical documentation processing, and report generation.
Qwen3-1.7B is a balanced model with 1.7 billion parameters and a 32K token context window, optimized for basic enterprise applications. It delivers high-quality dialogue and document analysis with moderate resource requirements, making it ideal for business chatbots and customer service automation systems.
Qwen3-0.6B is an ultra-compact language model with 600 million parameters and a 32K token context window, optimized for mobile devices and edge computing. The model delivers fast inference with minimal resource consumption and is ideal for IoT applications.
Qwen3-235B-A22B is the flagship open-source MoE model with 235 billion total parameters (22 billion active) and a context length of 128K tokens, delivering quality on par with the best proprietary models. The model is designed for mission-critical government systems, fundamental research, and flagship products where the highest level of modern AI quality is required.
Qwen3-30B-A3B is an advanced MoE (Mixture of Experts) model with a hybrid architecture that allows enabling or disabling reasoning mode as needed for flexible handling of tasks of varying complexity. With 30.5 billion parameters and dynamic activation of only 3.3 billion per token, along with support for context lengths of up to 128K, the model combines the quality of a large language model with the speed and efficiency of a smaller one.
GLM-4-32B-0414 is a powerful model with 32 billion parameters, trained on 15 TB of high-quality data. In terms of performance, it is comparable to leading models such as GPT-4o and DeepSeek-V3-0324, particularly in programming tasks, while remaining lightweight for easy local deployment.
GLM-Z1-9B-0414 is a compact reasoning model with 9.4 billion parameters. Despite its relatively small size, it demonstrates impressive step-by-step reasoning capabilities when performing general simple tasks. Thanks to an excellent balance between efficiency and performance, it is ideally suited for deployment in resource-constrained environments.
GLM-Z1-32B-0414 is a specialized reasoning model with 32B parameters and a 32K context length, trained through extended reinforcement learning (RL) to solve complex mathematical and logical problems. It is ideally suited for educational platforms, scientific research, and the development of systems requiring step-by-step analysis and solution justification.
GLM-Z1-Rumination-32B-0414 is a reasoning-capable model with 32 billion parameters, specifically trained to solve complex research and analytical tasks, with the ability to use external search. It excels at engaging in prolonged deliberation, allowing it to effectively handle multi-step assignments.
Llama 4 Scout is a model with native multimodality and a context window of up to 10 million tokens, while running on a single GPU. It is ideal for analyzing large text arrays and quickly extracting information from images.
Llama 4 Maverick - supports context windowing up to 1 million tokens, native multimodality and demonstrates high speed and efficiency due to the combination of 128 experts and 400 billion parameters in the architecture. The model is well suited for programming and technical documentation tasks.
DeepSeek-V3 0324 is an enhanced version of DeepSeek's powerful and popular MoE model with 685 billion parameters. Demonstrates exceptional quality, deep answer mining, and outstanding erudition in tasks ranging from analyzing complex legal documents to generating executable program code from scratch.