LFM2 is a family of hybrid models designed for on-device deployment. LFM2-24B-A2B is the largest model in the family, scaling the architecture to 24 billion parameters while keeping inference efficient.
919.3K Pulls 6 Tags Updated 1 week ago
Qwen 3.5 is a family of open-source multimodal models that delivers exceptional utility and performance.
677K Pulls 30 Tags Updated 3 days ago
MiniMax-M2.5 is a state-of-the-art large language model designed for real-world productivity and coding tasks.
88.4K Pulls 1 Tag Updated 3 weeks ago
A strong reasoning and agentic model from Z.ai with 744B total parameters (40B active), built for complex systems engineering and long-horizon tasks.
81.7K Pulls 1 Tag Updated 3 weeks ago
Qwen3-Coder-Next is a coding-focused language model from Alibaba's Qwen team, optimized for agentic coding workflows and local development.
735.5K Pulls 4 Tags Updated 4 weeks ago
LFM2.5 is a new family of hybrid models designed for on-device deployment.
940.4K Pulls 5 Tags Updated 1 month ago
A new collection of open translation models built on Gemma 3, helping people communicate across 55 languages.
506.9K Pulls 13 Tags Updated 1 month ago
The most powerful vision-language model in the Qwen model family to date.
1.8M Pulls 59 Tags Updated 4 months ago
As the strongest model in the 30B class, GLM-4.7-Flash offers a new option for lightweight deployment that balances performance and efficiency.
376.6K Pulls 4 Tags Updated 1 month ago
Building upon the foundational models of the Qwen3 series, Qwen3 Embedding provides a comprehensive range of text embeddings models in various sizes
1.1M Pulls 12 Tags Updated 5 months ago
The Ministral 3 family is designed for edge deployment, capable of running on a wide range of hardware.
536.2K Pulls 16 Tags Updated 2 months ago
Granite 4 features improved instruction following (IF) and tool-calling capabilities, making them more effective in enterprise applications.
768.7K Pulls 17 Tags Updated 4 months ago
The first installment in the Qwen3-Next series with strong performance in terms of both parameter efficiency and inference speed.
358.1K Pulls 10 Tags Updated 2 months ago
Rnj-1 is a family of 8B parameter open-weight, dense models trained from scratch by Essential AI, optimized for code and STEM with capabilities on par with SOTA open-weight models.
336.4K Pulls 6 Tags Updated 2 months ago
Kimi K2.5 is an open-source, native multimodal agentic model that seamlessly integrates vision and language understanding with advanced agentic capabilities, instant and thinking modes, as well as conversational and agentic paradigms.
121.5K Pulls 1 Tag Updated 1 month ago
DeepSeek-OCR is a vision-language model that can perform token-efficient OCR.
320.9K Pulls 3 Tags Updated 3 months ago
Nemotron 3 Nano - A new Standard for Efficient, Open, and Intelligent Agentic Models
194.3K Pulls 6 Tags Updated 2 months ago
24B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.
211K Pulls 6 Tags Updated 2 months ago
Olmo is a series of Open language models designed to enable the science of language models. These models are pre-trained on the Dolma 3 dataset and post-trained on the Dolci datasets.
182.6K Pulls 15 Tags Updated 2 months ago
GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture.
61.8K Pulls 3 Tags Updated 1 month ago
111.4K Pulls 10 Tags Updated 2 months ago
123B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.
100K Pulls 6 Tags Updated 2 months ago
FunctionGemma is a specialized version of Google's Gemma 3 270M model fine-tuned explicitly for function calling.
79.6K Pulls 4 Tags Updated 2 months ago
Gemini 3 Flash offers frontier intelligence built for speed at a fraction of the cost.
73.1K Pulls 2 Tags Updated 2 months ago
Advancing the Coding Capability
59.3K Pulls 1 Tag Updated 2 months ago
The Cogito v2.1 LLMs are instruction tuned generative models. All models are released under MIT license for commercial use.
82.1K Pulls 6 Tags Updated 3 months ago
nomic-embed-text-v2-moe is a multilingual MoE text embedding model that excels at multilingual retrieval.
51.8K Pulls 1 Tag Updated 2 months ago
gpt-oss-safeguard-20b and gpt-oss-safeguard-120b are safety reasoning models built-upon gpt-oss
79.4K Pulls 3 Tags Updated 4 months ago
MiniMax M2 is a high-efficiency large language model built for coding and agentic workflows.
71.1K Pulls 1 Tag Updated 4 months ago
Advanced agentic, reasoning and coding capabilities.
77.8K Pulls 1 Tag Updated 4 months ago
DeepSeek-V3.2, a model that harmonizes high computational efficiency with superior reasoning and agent performance.
41K Pulls 1 Tag Updated 2 months ago
Exceptional multilingual capabilities to elevate code engineering
21.7K Pulls 1 Tag Updated 2 months ago
Kimi K2 Thinking, Moonshot AI's best open-source thinking model.
34.3K Pulls 1 Tag Updated 3 months ago
A state-of-the-art mixture-of-experts (MoE) language model. Kimi K2-Instruct-0905 demonstrates significant improvements in performance on public benchmarks and real-world coding agent tasks.
42.4K Pulls 1 Tag Updated 5 months ago
A general-purpose multimodal mixture-of-experts model for production-grade tasks and enterprise workloads.
22.4K Pulls 1 Tag Updated 3 months ago
The current, most capable model that runs on a single GPU.
32.7M Pulls 29 Tags Updated 3 months ago
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.
22.8M Pulls 58 Tags Updated 4 months ago
OpenAI’s open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases.
7.4M Pulls 5 Tags Updated 4 months ago
Alibaba's performant long context models for agentic and coding tasks.
3.4M Pulls 10 Tags Updated 5 months ago
An update to Mistral Small that improves on function calling, instruction following, and less repetition errors.
1.3M Pulls 5 Tags Updated 8 months ago