The NVIDIA RTX 3060 Laptop GPU is a consumer-grade GPU built on the Ampere architecture, featuring 6 GB of memory with 336 GB/s of memory bandwidth and a 115 W TDP. Currently available from 1 provider starting at $0.03/GPU/hour, with a market median of $0.04/GPU/hour across 5 configurations.
Hardware specifications
Same across all providers
Median price across all providers
1 offering from 1 providers
Sorted by price ascending. Compare configurations side-by-side.
LLMs that fit on the NVIDIA RTX 3060 Laptop GPU
Estimated VRAM for popular open-weight LLMs at 16-bit (FP16/BF16), 8-bit (FP8/INT8) and 4-bit (INT4/MXFP4) precision, against 6 GB per GPU. Showing the 12 largest that fit.
| Model | Parameters | 16-bit | 8-bit | 4-bit |
|---|---|---|---|---|
| Qwen3.6 35B-A3B Alibaba | 35B 3B active | 86 GB Too large | 47 GB Barely fits on 8× | 28 GB 8× · 20 GB free |
| Qwen3 32B Alibaba | 32.8B | 79 GB Too large | 43 GB 8× · 4.8 GB free | 25 GB 8× · 23 GB free |
| Gemma 4 31B Google | 31.3B | 75 GB Too large | 41 GB 8× · 6.7 GB free | 24 GB 8× · 24 GB free |
| GLM-4.7 Flash Z.ai | 31.2B | 75 GB Too large | 41 GB 8× · 6.8 GB free | 24 GB 8× · 24 GB free |
| Qwen3.6 27B Alibaba | 27.8B | 67 GB Too large | 37 GB 8× · 11 GB free | 21 GB 8× · 27 GB free |
| Mistral Small 3.2 24B Mistral AI | 24B | 58 GB Too large | 32 GB 8× · 16 GB free | 18 GB 8× · 30 GB free |
| gpt-oss-20b OpenAI | 21B 3.6B active | — Not released | — Not released | 17 GB 8× · 31 GB free |
| Gemma 3 12B Google | 12.2B | 29 GB 8× · 19 GB free | 16 GB 8× · 32 GB free | 9.4 GB 2× · 2.6 GB free |
| Granite 4.1 8B IBM | 8.8B | 21 GB 8× · 27 GB free | 12 GB Barely fits on 2× | 6.8 GB 2× · 5.2 GB free or 1× · 6% spare |
| Qwen3 8B Alibaba | 8.2B | 20 GB 8× · 28 GB free | 11 GB 2× · 1.2 GB free | 6.3 GB 2× · 5.7 GB free or 1× · 14% spare |
| Llama 3.1 8B Instruct Meta | 8B | 19 GB 8× · 29 GB free | 11 GB 2× · 1.4 GB free | 6.2 GB 2× · 5.8 GB free or 1× · 16% spare |
| Mistral 7B Instruct v0.3 Mistral AI | 7.2B | 17 GB 8× · 31 GB free | 9.6 GB 2× · 2.4 GB free | 5.6 GB Barely fits on 1× |
Estimates: model weights plus 20% for KV cache and runtime overhead, at short context lengths. Long contexts and large batches need more. Rows marked “tight” hold the weights but not that full margin. A ×N figure is the total VRAM across N of these GPUs and makes no claim about interconnect throughput. How we estimate this
Want more from this page?
See incorrect data?