DeepSeek-OCR is a vision-language model that can perform token-efficient OCR.
442.6K Pulls 3 Tags Updated 5 months ago
The Cogito v2.1 LLMs are instruction tuned generative models. All models are released under MIT license for commercial use.
184K Pulls 6 Tags Updated 5 months ago
Kimi K2 Thinking, Moonshot AI's best open-source thinking model.
905.8K Pulls 1 Tag Updated 6 months ago
MiniMax M2 is a high-efficiency large language model built for coding and agentic workflows.
951K Pulls 1 Tag Updated 6 months ago
gpt-oss-safeguard-20b and gpt-oss-safeguard-120b are safety reasoning models built-upon gpt-oss
139.4K Pulls 3 Tags Updated 6 months ago
Advanced agentic, reasoning and coding capabilities.
955.5K Pulls 1 Tag Updated 7 months ago
The most powerful vision-language model in the Qwen model family to date.
3.7M Pulls 59 Tags Updated 6 months ago
Granite 4 features improved instruction following (IF) and tool-calling capabilities, making them more effective in enterprise applications.
1.2M Pulls 17 Tags Updated 6 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.
66.6K Pulls 1 Tag Updated 7 months ago
Building upon the foundational models of the Qwen3 series, Qwen3 Embedding provides a comprehensive range of text embeddings models in various sizes
1.9M Pulls 12 Tags Updated 7 months ago
EmbeddingGemma is a 300M parameter embedding model from Google.
1.2M Pulls 5 Tags Updated 8 months ago
DeepSeek-V3.1-Terminus is a hybrid model that supports both thinking mode and non-thinking mode.
676.8K Pulls 8 Tags Updated 7 months ago
OpenAI’s open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases.
9.5M Pulls 5 Tags Updated 7 months ago
Alibaba's performant long context models for agentic and coding tasks.
5.4M Pulls 10 Tags Updated 7 months ago
An update to Mistral Small that improves on function calling, instruction following, and less repetition errors.
2M Pulls 5 Tags Updated 10 months ago
Gemma 3n models are designed for efficient execution on everyday devices such as laptops, tablets or phones.
1.6M Pulls 9 Tags Updated 10 months ago
Magistral is a small, efficient reasoning model with 24B parameters.
1.4M Pulls 5 Tags Updated 11 months ago
Devstral: the best open source model for coding agents
939.3K Pulls 5 Tags Updated 10 months ago
Flagship vision-language model of Qwen and also a significant leap from the previous Qwen2-VL.
1.9M Pulls 17 Tags Updated 11 months ago
Phi 4 reasoning and reasoning plus are 14-billion parameter open-weight reasoning models that rival much larger models on complex reasoning tasks.
1.6M Pulls 9 Tags Updated 1 year ago