Quickstart
Pick a category and your system size — we show the best local models with ready-to-copy setup commands.
Category
Best local models for code generation, summarization, and general reasoning. Benchmarked on 10-task LLM coding eval.
SmolLM3 3BScore: 93.3%1.8 GB
Ollama
ollama run smollm3-3bllama.cpp
llama-cli -m smollm3-3b.gguf -p "Your prompt" -ngl 99Qwen2.5 1.5BScore: 85%0.9 GB
Ollama
ollama run qwen2.5-1.5bllama.cpp
llama-cli -m qwen2.5-1.5b.gguf -p "Your prompt" -ngl 99Qwen2.5 3BScore: 85%1.8 GB
Ollama
ollama run qwen2.5-3bllama.cpp
llama-cli -m qwen2.5-3b.gguf -p "Your prompt" -ngl 99Granite 3.2 2BScore: 82.5%1.5 GB
Ollama
ollama run granite-3.2-2bllama.cpp
llama-cli -m granite-3.2-2b.gguf -p "Your prompt" -ngl 99Ministral 3BScore: 81.7%2.0 GB
Ollama
ollama run ministral-3bllama.cpp
llama-cli -m ministral-3b.gguf -p "Your prompt" -ngl 99Falcon3-3B-Instruct-4bitScore: 79%1.7 GB
Ollama
ollama run falcon3-3b-instruct-4bitllama.cpp
llama-cli -m falcon3-3b-instruct-4bit.gguf -p "Your prompt" -ngl 99Qwen2.5 0.5BScore: 74.2%0.4 GB
Ollama
ollama run qwen2.5-0.5bllama.cpp
llama-cli -m qwen2.5-0.5b.gguf -p "Your prompt" -ngl 99Llama 3.2 1BScore: 73.3%0.8 GB
Ollama
ollama run llama-3.2-1bllama.cpp
llama-cli -m llama-3.2-1b.gguf -p "Your prompt" -ngl 99SmolLM2 1.7BScore: 70.8%1.0 GB
Ollama
ollama run smollm2-1.7bllama.cpp
llama-cli -m smollm2-1.7b.gguf -p "Your prompt" -ngl 99Falcon3-1B-Instruct-4bitScore: 62%0.9 GB
Ollama
ollama run falcon3-1b-instruct-4bitllama.cpp
llama-cli -m falcon3-1b-instruct-4bit.gguf -p "Your prompt" -ngl 99DeepSeek-R1 1.5BScore: 27.5%1.0 GB
Ollama
ollama run deepseek-r1-1.5bllama.cpp
llama-cli -m deepseek-r1-1.5b.gguf -p "Your prompt" -ngl 99Qwen3.5 0.8BScore: 26%0.5 GB
Ollama
ollama run qwen3.5-0.8bllama.cpp
llama-cli -m qwen3.5-0.8b.gguf -p "Your prompt" -ngl 99Top Cloud APIs (coding)
| Model | Score | Provider |
|---|---|---|
| IBM Granite 4.1 8B | 90% | openrouter.ai |
| Nemotron 3 Nano 30B A3B | 90% | openrouter.ai |
| Codestral 2508 | 90% | openrouter.ai |
| MiniMax M2 Her | 90% | openrouter.ai |
| DeepSeek Chat | 90% | openrouter.ai |
| Qwen3 Coder 30B A3B | 90% | openrouter.ai |
| Mistral Large 2411 | 90% | openrouter.ai |
| DeepSeek Chat V3-0324 | 90% | openrouter.ai |