跳至主要內容
kitlocal.dev

Ollama 模型參考

本機執行

100% 在你的瀏覽器中處理。輸入與輸出絕不會離開你的裝置。

瀏覽 Ollama 模型資料庫,包含大小、功能與使用資訊。

llama3.1

Meta's latest Llama model. Great general-purpose performance.

參數規模: 8B/70B/405B大小: 4.7GB/40GB/231GB上下文長度: 128K
generalcodechat
$ ollama run llama3.1

llama3.2

Lightweight Llama models optimized for edge and mobile.

參數規模: 1B/3B大小: 1.3GB/2.0GB上下文長度: 128K
edgesmall
$ ollama run llama3.2

mistral

Mistral 7B - efficient model with strong performance.

參數規模: 7B大小: 4.1GB上下文長度: 32K
generalfast
$ ollama run mistral

mixtral

Mixture of Experts model. Sparse activation for efficiency.

參數規模: 8x7B/8x22B大小: 26GB/80GB上下文長度: 64K
generalMoE
$ ollama run mixtral

codellama

Code-specialized Llama. Supports code completion and infilling.

參數規模: 7B/13B/34B/70B大小: 3.8GB/7.4GB/19GB/40GB上下文長度: 16K
codecompletion
$ ollama run codellama

deepseek-coder-v2

DeepSeek Coder v2. State-of-the-art coding model.

參數規模: 16B/236B大小: 8.9GB/133GB上下文長度: 128K
codemath
$ ollama run deepseek-coder-v2

phi3

Microsoft Phi-3. Small but capable reasoning model.

參數規模: 3.8B/14B大小: 2.2GB/7.9GB上下文長度: 128K
reasoningsmall
$ ollama run phi3

gemma2

Google Gemma 2. Efficient open model family.

參數規模: 2B/9B/27B大小: 1.6GB/5.4GB/16GB上下文長度: 8K
generalgoogle
$ ollama run gemma2

qwen2.5

Alibaba Qwen 2.5. Strong multilingual and coding.

參數規模: 0.5B-72B大小: 0.4-41GB上下文長度: 128K
multilingualcode
$ ollama run qwen2.5

nomic-embed-text

Text embedding model for RAG and semantic search.

參數規模: 137M大小: 274MB上下文長度: 8K
embeddingRAG
$ ollama run nomic-embed-text