model,family,params,quantization,minRamGb,estimatedLoadGb,runsLocally,openWeights,ggufDiy,runtimes,bestFor,ollamaCommand
Qwen2.5 1.5B Instruct,Qwen,1.5,Q4_K_M,4,1.5,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Translation",ollama run qwen2.5:1.5b-instruct-q4_K_M
Qwen2.5 3B Instruct,Qwen,3,Q4_K_M,6,2.5,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run qwen2.5:3b-instruct-q4_K_M
Qwen2.5 7B Instruct,Qwen,7,Q4_K_M,10,5.5,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run qwen2.5:7b-instruct-q4_K_M
Qwen2.5 14B Instruct,Qwen,14,Q4_K_M,20,11,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Chat",ollama run qwen2.5:14b-instruct-q4_K_M
Llama 3.2 3B Instruct,Llama,3,Q4_K_M,6,2.5,true,true,false,ollama|llama.cpp|lm-studio,Chat,ollama run llama3.2:3b-instruct-q4_K_M
Llama 3.1 8B Instruct,Llama,8,Q4_K_M,12,6.5,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run llama3.1:8b-instruct-q4_K_M
Llama 3.1 8B Instruct (Q5),Llama,8,Q5_K_M,14,8,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run llama3.1:8b-instruct-q5_K_M
Llama 3.1 70B Instruct,Llama,70,Q4_K_M,48,42,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding",ollama run llama3.1:70b-instruct-q4_K_M
Mistral 7B Instruct,Mistral,7,Q4_K_M,10,5.5,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run mistral:7b-instruct-q4_K_M
Mixtral 8x7B Instruct,Mistral,46.7,Q4_K_M,36,30,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Quality",ollama run mixtral:8x7b
Mistral Nemo 12B,Mistral,12,Q4_K_M,18,9.5,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Translation",ollama run mistral-nemo:12b
Gemma 2 2B Instruct,Gemma,2,Q4_K_M,5,1.8,true,true,false,ollama|llama.cpp|lm-studio,Chat,ollama run gemma2:2b-instruct-q4_K_M
Gemma 2 9B Instruct,Gemma,9,Q4_K_M,14,7,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run gemma2:9b-instruct-q4_K_M
Gemma 2 27B Instruct,Gemma,27,Q4_K_M,28,21,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding",ollama run gemma2:27b-instruct-q4_K_M
Phi-3 Mini 3.8B,Phi,3.8,Q4_K_M,7,3.2,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Chat",ollama run phi3:mini
Phi-3 Medium 14B,Phi,14,Q4_K_M,20,11,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Quality",ollama run phi3:medium
Phi-4 14B,Phi,14,Q4_K_M,22,11.5,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Quality",ollama run phi4:14b-q4_K_M
Qwen2.5 Coder 7B,Qwen,7,Q4_K_M,10,5.5,true,true,false,ollama|llama.cpp|lm-studio,Coding,ollama run qwen2.5-coder:7b
Qwen2.5 Coder 14B,Qwen,14,Q4_K_M,22,11,true,true,false,ollama|llama.cpp|lm-studio,Coding,ollama run qwen2.5-coder:14b
Llama 3.2 1B Instruct,Llama,1,Q4_K_M,3,1,true,true,false,ollama|llama.cpp|lm-studio,Chat,ollama run llama3.2:1b-instruct-q4_K_M
Mistral Small 22B,Mistral,22,Q4_K_M,26,17,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Quality",ollama run mistral-small:22b
Kimi K2 Instruct,Kimi,1000,API,0,0,false,true,false,,"Quality, Coding",
Claude 3.5 Sonnet,Claude,,API,0,0,false,false,false,,"Coding, Chat",
Claude 3.7 Sonnet,Claude,,API,0,0,false,false,false,,"Coding, Quality",
Claude 3 Opus,Claude,,API,0,0,false,false,false,,"Quality, Analysis",
Claude 4 Opus,Claude,,API,0,0,false,false,false,,"Quality, Coding",
Qwen2.5 0.5B Instruct,Qwen,0.5,Q4_K_M,2,0.8,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Mobile",ollama run qwen2.5:0.5b-instruct-q4_K_M
Gemma 3 1B Instruct,Gemma,1,Q4_K_M,3,1,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Mobile",ollama run gemma3:1b
Phi-4 Mini 3.8B,Phi,3.8,Q4_K_M,7,3.2,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Chat",ollama run phi4-mini:3.8b
SmolLM2 360M,SmolLM,0.36,Q4_K_M,1,0.5,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Embedded",ollama run smollm2:360m
Llama 3.3 70B Instruct,Llama,70,Q4_K_M,48,42,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding",ollama run llama3.3:70b-instruct-q4_K_M
Llama 3.1 405B Instruct,Llama,405,Q4_K_M,256,243,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Reasoning, Coding",ollama run llama3.1:405b-instruct-q4_K_M
DeepSeek-R1 671B,DeepSeek,671,Q4_K_M,400,380,true,true,false,ollama|llama.cpp|lm-studio,"Reasoning, Coding",ollama run deepseek-r1:671b-q4_K_M
Qwen3 8B,Qwen,8,Q4_K_M,12,6.5,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run qwen3:8b-q4_K_M
Qwen3 14B,Qwen,14,Q4_K_M,20,11,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Quality",ollama run qwen3:14b-q4_K_M
Qwen3 30B,Qwen,30,Q4_K_M,28,22,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding",ollama run qwen3:30b
Gemma 3 4B Instruct,Gemma,4,Q4_K_M,8,3.5,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run gemma3:4b
Gemma 3 12B Instruct,Gemma,12,Q4_K_M,18,9.5,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Quality",ollama run gemma3:12b
Gemma 3 27B Instruct,Gemma,27,Q4_K_M,28,21,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding",ollama run gemma3:27b
DeepSeek-R1 Distill Qwen 7B,DeepSeek,7,Q4_K_M,10,5.5,true,true,false,ollama|llama.cpp|lm-studio,"Reasoning, Coding",ollama run deepseek-r1:7b
DeepSeek-R1 Distill Qwen 14B,DeepSeek,14,Q4_K_M,22,11,true,true,false,ollama|llama.cpp|lm-studio,"Reasoning, Quality",ollama run deepseek-r1:14b
DeepSeek-R1 Distill Llama 70B,DeepSeek,70,Q4_K_M,48,42,true,true,false,ollama|llama.cpp|lm-studio,"Reasoning, Quality",ollama run deepseek-r1:70b
GPT-4o,OpenAI,,API,0,0,false,false,false,,"Quality, Coding",
GPT-4o mini,OpenAI,,API,0,0,false,false,false,,"Chat, Speed",
Claude 4 Sonnet,Claude,,API,0,0,false,false,false,,"Coding, Quality",
Gemini 2.5 Pro,Google,,API,0,0,false,false,false,,"Quality, Analysis",
Gemini 2.5 Flash,Google,,API,0,0,false,false,false,,"Speed, Chat",
DeepSeek-V3,DeepSeek,671,API,0,0,false,true,false,,"Quality, Coding",
GLM-5,Zhipu,744,API,0,0,false,true,false,,"Reasoning, Quality",
GLM-4 Plus,Zhipu,,API,0,0,false,false,false,,"Quality, Analysis",
Qwen3 235B A22B,Qwen,235,Q4_K_M,192,130,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Reasoning",ollama run qwen3:235b-a22b-q4_K_M
Mistral Small 3.1,Mistral,24,Q4_K_M,24,15,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run mistral-small3.1:24b
DeepSeek-V3-0324,DeepSeek,671,API,0,0,false,true,false,,"Quality, Coding",
DeepSeek-R1,DeepSeek,671,API,0,0,false,true,false,,"Reasoning, Quality",
Qwen3.5 0.8B Instruct,Qwen,0.8,Q4_K_M,2,0.8,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Mobile",ollama run qwen3.5:0.8b
Qwen3.5 2B Instruct,Qwen,2,Q4_K_M,4,1.8,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Edge tasks",ollama run qwen3.5:2b
Qwen3.5 4B Instruct,Qwen,4,Q4_K_M,8,3.5,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Agents, Multimodal",ollama run qwen3.5:4b
Qwen3.5 9B Instruct,Qwen,9,Q4_K_M,14,7,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding, Reasoning",ollama run qwen3.5:9b
Qwen3.5 35B-A3B Instruct,Qwen,35,Q4_K_M,24,20,true,true,false,ollama|llama.cpp|lm-studio,"Reasoning, Coding, Agent scenarios",ollama run qwen3.5:35b-a3b
Qwen3.5 27B Instruct,Qwen,27,Q4_K_M,20,16,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding, Complex reasoning",ollama run qwen3.5:27b
Qwen3.5 122B-A10B Instruct,Qwen,122,Q4_K_M,96,72,true,true,false,ollama|llama.cpp|lm-studio,"Frontier-level reasoning, Complex tasks",ollama run qwen3.5:122b-a10b
LFM2 24B-A2B Instruct,LFM2,24,Q4_K_M,16,14,true,true,false,ollama|llama.cpp|lm-studio,"Local AI agents, privacy-first tool calling, MCP workflows",ollama run lfm2:24b-a2b
LFM2.5 8B-A1B,LFM2,8.3,Q4_K_M,10,5.5,true,true,false,ollama|llama.cpp|lm-studio,"On-device agents, tool calling, multilingual chat",ollama run lfm2.5:8b-a1b-q4_K_M
Granite 4.1 3B Instruct,Granite,3,Q4_K_M,6,2,true,true,false,ollama|llama.cpp|lm-studio,"Lightweight chat, classification, edge tasks",ollama run granite4.1:3b
Granite 4.1 8B Instruct,Granite,8,Q4_K_M,12,5.5,true,true,false,ollama|llama.cpp|lm-studio,"Enterprise assistant, tool calling, instruction following",ollama run granite4.1:8b
Qwen3.6 27B,Qwen,27,Q4_K_M,24,18,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Quality, Long context",ollama run qwen3.6:27b
Qwen3.6 35B-A3B,Qwen,35,Q4_K_M,24,22,true,true,false,ollama|llama.cpp|lm-studio,"Reasoning, Coding, Agents",ollama run qwen3.6:35b-a3b
Gemma 4 31B,Gemma,31,Q4_K_M,32,20,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding, Multimodal",ollama run gemma4:31b
Gemma 4 26B-A4B,Gemma,26,Q4_K_M,24,16,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding, Multimodal",ollama run gemma4:26b
Gemma 4 E4B,Gemma,4.5,Q4_K_M,8,4,true,true,false,ollama|llama.cpp|lm-studio,"On-device, Mobile, Chat",ollama run gemma4:e4b
Gemma 4 E2B,Gemma,2.3,Q4_K_M,4,2.3,true,true,false,ollama|llama.cpp|lm-studio,"IoT, Mobile, Edge",ollama run gemma4:e2b
Llama 4 Scout,Llama,109,Q4_K_M,80,67,true,true,false,ollama|llama.cpp|lm-studio,"Long context, Quality, Multimodal",ollama run llama4:scout
Llama 4 Maverick,Llama,400,Q4_K_M,256,245,true,true,false,ollama|llama.cpp|lm-studio,"Frontier quality, Long context",ollama run llama4:maverick
Mistral Medium 3.5,Mistral,128,API,0,0,false,true,true,,"Coding, Quality, Tool calling",
DeepSeek V4 Flash,DeepSeek,284,API,0,0,false,true,true,,"Reasoning, Coding, Long context",
DeepSeek V4 Pro,DeepSeek,1600,API,0,0,false,true,false,,"Frontier reasoning, Coding",
Kimi K2.6,Kimi,1000,API,0,0,false,true,false,,"Coding, Agentic tasks",
GLM-5.1,Zhipu,744,API,0,0,false,true,false,,"Coding, Reasoning, Long context",
Claude Opus 4.7,Claude,,API,0,0,false,false,false,,"Coding, Quality, Reasoning",
Claude Opus 4.8,Claude,,API,0,0,false,false,false,,"Coding, Agentic, Reasoning",
GPT-5.5,OpenAI,,API,0,0,false,false,false,,"Quality, Reasoning, Multimodal",
Gemini 3.1 Pro,Google,,API,0,0,false,false,false,,"Multimodal, Long context, Quality",
Grok 4.3,xAI,,API,0,0,false,false,false,,"Reasoning, Real-time data, Coding",
Gemma 4 12B,Gemma,12,Q4_K_M,16,8,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding, Multimodal",ollama run gemma4:12b
Claude Fable 5,Claude,,API,0,0,false,false,false,,"Coding, Agentic, Reasoning",
GLM-5.2,Zhipu,,API,0,0,false,false,false,,"Coding, Reasoning, Agentic",
Kimi K2.7-Code,Kimi,1000,API,0,0,false,true,false,,"Coding, Agentic tasks",
MiniMax M3,MiniMax,428,API,0,0,false,true,false,,"Coding, Agentic, Long context",
Qwen3.7-Plus,Qwen,,API,0,0,false,false,false,,"Multimodal, Agentic, Vision",
NVIDIA Nemotron 3 Ultra,Nemotron,550,API,0,0,false,true,false,,"Reasoning, Agentic, Long context",
Xiaomi MiMo-V2-Flash,MiMo,309,API,0,0,false,true,true,,"Reasoning, Coding, Agentic tasks",
NVIDIA Nemotron Cascade 2 30B-A3B,Nemotron,30,Q6_K,32,24,true,true,false,ollama|llama.cpp|lm-studio,"Reasoning, Math, Agentic tasks",ollama run nemotron-cascade-2:30b
Poolside Laguna XS.2,Laguna,33,Q4_K_M,32,23,true,true,false,ollama|llama.cpp|lm-studio,"Agentic coding, long-horizon software engineering",ollama run laguna-xs.2:q4_K_M
Cohere North Mini Code,North,30,Q4_K_M,24,19,true,true,false,ollama|llama.cpp|lm-studio,"Agentic coding, repo-scale software engineering",ollama run north-mini-code-1.0:q4_K_M
GPT-OSS 120B,GPT-OSS,117,MXFP4,96,65.4,true,true,false,ollama|llama.cpp|lm-studio,"Reasoning, Coding, Agents",ollama run gpt-oss:120b
GPT-OSS 20B,GPT-OSS,21,MXFP4,24,13.8,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding, Reasoning",ollama run gpt-oss:20b
Qwen3-Next 80B-A3B,Qwen,80,Q4_K_M,72,50.4,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding, Long Context",ollama run qwen3-next:80b
Qwen3.6 35B-A3B (Q8),Qwen,35,Q8_0,64,38.7,true,true,false,ollama|llama.cpp|lm-studio,"Reasoning, Coding, Agents",ollama run qwen3.6:35b-a3b-q8_0
Qwen3.5 35B-A3B Instruct (Q8),Qwen,35,Q8_0,64,38.7,true,true,false,ollama|llama.cpp|lm-studio,"Reasoning, Coding, Agent scenarios",ollama run qwen3.5:35b-a3b-q8_0
Qwen3.5 9B Instruct (Q8),Qwen,9,Q8_0,16,10.7,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding, Reasoning",ollama run qwen3.5:9b-q8_0
Qwen3.6 27B (Q8),Qwen,27,Q8_0,48,30,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Quality, Long context",ollama run qwen3.6:27b-q8_0
Gemma 4 12B (Q8),Gemma,12,Q8_0,24,12.8,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding, Multimodal",ollama run gemma4:12b-it-q8_0
Gemma 4 26B-A4B (Q8),Gemma,26,Q8_0,48,28.1,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding, Multimodal",ollama run gemma4:26b-a4b-it-q8_0
Qwen3-Next 80B-A3B (Q8),Qwen,80,Q8_0,128,84.8,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding, Long Context",ollama run qwen3-next:80b-a3b-instruct-q8_0
Qwen3 14B (Q8),Qwen,14,Q8_0,24,15.9,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Quality",ollama run qwen3:14b-q8_0
Llama 3.3 70B Instruct (Q8),Llama,70,Q8_0,128,75,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding",ollama run llama3.3:70b-instruct-q8_0
Llama 3.3 70B Instruct (Q6),Llama,70,Q6_K,96,57.9,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding",ollama run llama3.3:70b-instruct-q6_K