Hardware → models
RTX 3060 12 GB
Start with Qwen3 14B at Q4_K_M for general chat. For a coding-focused option, try Qwen2.5-Coder 14B.
Models to start with
Planning estimates12 GB dedicated VRAM · 4,000 tokens · 32 GB system RAM · full GPU residency
| Model / starting file | Why consider it | GPU memory estimate | Memory left¹ |
|---|---|---|---|
| Qwen3 14BQ4_K_M | General chat with a larger modelUses more memory than 8B. A larger model is not proof of better answers for your task. | 10.2 GB | 1.4 GB |
| Qwen2.5-Coder 14BQ4_K_M | Coding with a larger modelMore memory than the 7B version. Compare them on your own codebase. | 10.2 GB | 1.4 GB |
| Qwen3 8BQ4_K_M | General chat and mixed tasksA middle-sized option with more memory headroom than the 14B models. | 6.2 GB | 5.4 GB |
| Llama 3.2 3B InstructQ4_K_M | Small text assistantA smaller weight footprint. Check the exact Llama checkpoint and licence. | 3.7 GB | 7.9 GB |
From the published model guides here; the Finder covers the full catalogue. ¹ Memory left after the Finder’s reserve. Other applications can use this space. Larger models and higher precision do not guarantee better answers.
14B Q4 profiles fit at 4K in this estimate, but leave less headroom. gpt-oss-20b does not fit this full-GPU budget. CPU offloading changes the memory allocation and may slow generation.
Setup and common questions
Is this the right RTX 3060 version?
This page covers the desktop 12 GB RTX 3060. NVIDIA also lists an 8 GB version; the RTX 3060 Ti is a different 8 GB card. Check the exact card before selecting it in the Finder.
What should I check on my first run?
Load one complete model file at 4K context in an app with a working NVIDIA backend. Confirm GPU offload and watch memory during generation. If it runs out of memory, close other GPU workloads or reduce context.
Can system RAM make up for limited VRAM?
Partial CPU offload can help, but changes performance and needs enough system RAM. Operating-system shared GPU memory is not extra dedicated VRAM. A memory fit alone does not establish speed.
How are these recommendations chosen?
We use the main Finder’s balanced task selection and memory calculation, limited to models with published guides here. Task scores and speed proxies are planning inputs, not measured quality rankings.
GPU memory includes weights, context cache and a 0.9 GB runtime allowance, plus a 3% reserve (0.15–1.5 GB). Cache format and runtime buffers can change the fit. See the calculation assumptions.
Sources and related guides
Specification sources checked 2 October 2026.
These comparisons are planning estimates. No reviewed or independently reproduced measurements are available yet. About test evidence.