Ollama

gpt-oss

OpenAI’s open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases.

tools thinking cloud 20b 120b

9.7M Pulls 5 Tags Updated 7 months ago

qwen3-vl

The most powerful vision-language model in the Qwen model family to date.

vision tools thinking cloud 2b 4b 8b 30b 32b 235b

3.8M Pulls 59 Tags Updated 6 months ago

qwen3-coder

Alibaba's performant long context models for agentic and coding tasks.

tools cloud 30b 480b

5.5M Pulls 10 Tags Updated 7 months ago

kimi-k2-thinking

Kimi K2 Thinking, Moonshot AI's best open-source thinking model.

tools thinking cloud

1.7M Pulls 1 Tag Updated 6 months ago

minimax-m2

MiniMax M2 is a high-efficiency large language model built for coding and agentic workflows.

tools thinking cloud

1.8M Pulls 1 Tag Updated 6 months ago

glm-4.6

Advanced agentic, reasoning and coding capabilities.

tools thinking cloud

1.8M Pulls 1 Tag Updated 7 months ago

qwen3-embedding

Building upon the foundational models of the Qwen3 series, Qwen3 Embedding provides a comprehensive range of text embeddings models in various sizes

embedding 0.6b 4b 8b

1.9M Pulls 12 Tags Updated 7 months ago

mistral-small3.2

An update to Mistral Small that improves on function calling, instruction following, and less repetition errors.

vision tools 24b

2.1M Pulls 5 Tags Updated 11 months ago

granite4

Granite 4 features improved instruction following (IF) and tool-calling capabilities, making them more effective in enterprise applications.

tools 350m 1b 3b

1.2M Pulls 17 Tags Updated 6 months ago

embeddinggemma

EmbeddingGemma is a 300M parameter embedding model from Google.

embedding 300m

1.2M Pulls 5 Tags Updated 8 months ago

gemma3n

Gemma 3n models are designed for efficient execution on everyday devices such as laptops, tablets or phones.

e2b e4b

1.6M Pulls 9 Tags Updated 10 months ago

magistral

Magistral is a small, efficient reasoning model with 24B parameters.

tools thinking 24b

1.4M Pulls 5 Tags Updated 11 months ago

deepseek-ocr

DeepSeek-OCR is a vision-language model that can perform token-efficient OCR.

vision 3b

447.6K Pulls 3 Tags Updated 6 months ago

deepseek-v3.1

DeepSeek-V3.1-Terminus is a hybrid model that supports both thinking mode and non-thinking mode.

tools thinking cloud 671b

682.6K Pulls 8 Tags Updated 7 months ago

cogito-2.1

The Cogito v2.1 LLMs are instruction tuned generative models. All models are released under MIT license for commercial use.

cloud 671b

187.9K Pulls 6 Tags Updated 5 months ago

gpt-oss-safeguard

gpt-oss-safeguard-20b and gpt-oss-safeguard-120b are safety reasoning models built-upon gpt-oss

tools thinking 20b 120b

140.7K Pulls 3 Tags Updated 6 months ago

kimi-k2

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.

tools cloud

69.6K Pulls 1 Tag Updated 7 months ago

deepseek-r1

DeepSeek-R1 is a family of open reasoning models with performance approaching that of leading models, such as O3 and Gemini 2.5 Pro.

tools thinking 1.5b 7b 8b 14b 32b 70b 671b

85.8M Pulls 35 Tags Updated 10 months ago

llama3.1

Llama 3.1 is a new state-of-the-art model from Meta available in 8B, 70B and 405B parameter sizes.

tools 8b 70b 405b

114.6M Pulls 93 Tags Updated 1 year ago

llama3.2

Meta's Llama 3.2 goes small with 1B and 3B models.

tools 1b 3b

69.7M Pulls 63 Tags Updated 1 year ago