Kendi bilgisayarınızda veya sunucunuzda ücretsiz çalıştırabileceğiniz modeller. Katalog ollama.com kütüphanesiyle canlı senkrondur · Son senkron: 13.08.2026 22:48.
89 model
GLM-5.1 is our next-generation flagship model for agentic engineering, with significantly stronger coding capabilities than its predecessor. It achieves state-of-the-art performance on SWE-Bench Pro and leads GLM-5 by a wide margin.
⬇ 2.3M indirme · 0 varyant · 04.2026
An open 30B MoE model from NVIDIA with 3B activated parameters that delivers strong reasoning and agentic capabilities.
⬇ 140.9K indirme · 3 varyant · 03.2026
MiniMax's M2-series model for coding, agentic workflows, and professional productivity.
⬇ 2.3M indirme · 0 varyant · 03.2026
Nemotron-3-Nano is a new Standard for Efficient, Open, and Intelligent Agentic Models, now updated with a 4B parameter count model.
⬇ 686.6K indirme · 9 varyant · 03.2026
NVIDIA Nemotron 3 Super is a 120B open MoE model activating just 12B parameters to deliver maximum compute efficiency and accuracy for complex multi-agent applications.
⬇ 2.9M indirme · 7 varyant · 03.2026
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.
⬇ 1.1M indirme · 6 varyant · 02.2026
Qwen3-Coder-Next is a coding-focused language model from Alibaba's Qwen team, optimized for agentic coding workflows and local development.
⬇ 2M indirme · 3 varyant · 02.2026
GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture.
⬇ 7M indirme · 3 varyant · 02.2026
LFM2.5 is a new family of hybrid models designed for on-device deployment.
⬇ 1.3M indirme · 5 varyant · 01.2026
FunctionGemma is a specialized version of Google's Gemma 3 270M model fine-tuned explicitly for function calling.
⬇ 180K indirme · 4 varyant · 12.2025
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.
⬇ 284.5K indirme · 10 varyant · 12.2025
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.
⬇ 448.5K indirme · 15 varyant · 12.2025
24B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.
⬇ 935.1K indirme · 6 varyant · 12.2025
The Ministral 3 family is designed for edge deployment, capable of running on a wide range of hardware.
⬇ 1.4M indirme · 13 varyant · 12.2025
123B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.
⬇ 351.2K indirme · 5 varyant · 12.2025
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.
⬇ 501.7K indirme · 5 varyant · 12.2025
The first installment in the Qwen3-Next series with strong performance in terms of both parameter efficiency and inference speed.
⬇ 579.2K indirme · 9 varyant · 12.2025
A general-purpose multimodal mixture-of-experts model for production-grade tasks and enterprise workloads.
⬇ 90.1K indirme · 0 varyant · 12.2025
The most powerful vision-language model in the Qwen model family to date.
⬇ 5.2M indirme · 57 varyant · 10.2025
Granite 4 features improved instruction following (IF) and tool-calling capabilities, making them more effective in enterprise applications.
⬇ 1.4M indirme · 17 varyant · 10.2025
gpt-oss-safeguard-20b and gpt-oss-safeguard-120b are safety reasoning models built-upon gpt-oss
⬇ 150.8K indirme · 3 varyant · 10.2025
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.
⬇ 34.1M indirme · 58 varyant · 10.2025
OpenAI’s open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases.
⬇ 11.7M indirme · 5 varyant · 10.2025
DeepSeek-V3.1-Terminus is a hybrid model that supports both thinking mode and non-thinking mode.
⬇ 720.5K indirme · 7 varyant · 09.2025
Bu modeller Ollama ile tek komutla bilgisayarınızda çalışır (Windows, macOS, Linux). Verileriniz cihazınızdan çıkmaz — gizlilik gerektiren işler için idealdir.
ollama run … komutunu terminale yapıştırın