NVIDIA drivers with no limiter. Some files are stored using git LFS
-
Updated
Mar 15, 2021 - HLSL
#
3060
Here are 5 public repositories matching this topic...
Learning DeepSeek-R1 on low-end hardware(eg: 3060 12G) 在低端硬件（例如：3060 12G）上学习 DeepSeek-R1
-
Updated
Feb 9, 2025 - Python
A self-contained CUDA inference engine for LiquidAI/LFM2.5-8B-A1B (hybrid conv + GQA-attention MoE, 8.5B params, 1B active) targeting a single RTX 3060 (12 GB). No Python, no frameworks at runtime: a single .cu engine + a header-only byte-level BPE tokenizer.
c open-source ai cpp gpu cuda inference nvidia moe token rtx inference-engine fast-inference 3060 large-language-models llm liquidai lfm25
-
Updated
Jun 7, 2026 - Cuda
Add this topic to your repo
To associate your repository with the 3060 topic, visit your repo's landing page and select "manage topics."