Best local LLM for rtx 3080 12

A 14B dense model from Alibaba that matches Qwen 2.5 32B Base on STEM and code, with the same hybrid thinking system as the rest of the Qwen 3 family. The pragmatic sweet spot for a single 24GB GPU.

Microsoft's 14B reasoner that beats R1-Distill-Llama-70B on AIME and GPQA with 50x fewer parameters. MIT-licensed, English-first, with a 32K context.

DeepSeek's R1 reasoning distilled into Qwen 14B under MIT. AIME24 69.7 and MATH-500 93.9 — beats o1-mini on most reasoning benchmarks.

IBM's edge-class hybrid MoE with 7B total and only 1B active parameters — Apache 2.0 licensed and built for embedded and low-cost serving.

Microsoft's Phi-4 14B, trained on ultra-curated synthetic data with a heavy STEM bias. The 14B reasoning leader at the end of 2024.

Mistral AI and NVIDIA's co-developed 12B instruct model with 128k context, the Tekken tokenizer, and strong European multilingual coverage.

The 12B sweet spot of Google's Gemma 3 line — multimodal, 128K context, and 140 languages. Fits on a single consumer GPU with room for batching.