RTX 3060 12GB for local LLMs
10 measured results from 3 builds, each linked to its source.
| Memory | 12 GB per card |
|---|---|
| Memory bandwidth | 360 GB/s (memory bandwidth as listed on llamabench) |
| Typical EU price | ≈€429 new (Asus GeForce RTX 3060 V2 Dual 12 GB GDDR6, billiger.de, 2026-10-16) |
Can it run? Gemma 4 12B on RTX 3060 12GB LFM2.5 24B-A2B on RTX 3060 12GB Qwen3.8 27B on RTX 3060 12GB gpt-oss 20B on RTX 3060 12GB
All builds with the RTX 3060 12GB
| Build | GPU(s) | Memory GB (total) | Results |
|---|---|---|---|
| i7-12700K + RTX 5060 Ti 16GB + RTX 3060 12GB: Qwen3.8 27B from 21 to 51 tok/s (split, MTP, single-GPU) | RTX 5060 Ti 16GB + RTX 3060 12GB | 28 | 4 |
| Intel Core Ultra 265K + RTX 3060 | RTX 3060 | 12 | 3 |
| RTX 3060 12GB: 2026 models (llamabench community) | RTX 3060 | 12 | 3 |
All measured results
(MTP, DFlash, DSpark and similar draft-and-verify methods give much higher tok/s on predictable text and are not comparable 1:1 with plain decode)
| Model | Build | Ran on | Quant | Engine | tok/s | Details |
|---|---|---|---|---|---|---|
| Gemma 4 12B | RTX 3060 12GB: 2026 models (llamabench community) | RTX 3060 12GB | llama-bench tg128 (quant per result) | llama.cpp | 38.5 | |
| LFM2.5 24B-A2B | RTX 3060 12GB: 2026 models (llamabench community) | RTX 3060 12GB | llama-bench tg128 (quant per result) | llama.cpp | 125 | partly in system RAM |
| Qwen3.8 27B | i7-12700K + RTX 5060 Ti 16GB + RTX 3060 12GB: Qwen3.8 27B from 21 to 51 tok/s (split, MTP, single-GPU) | mixed / see build | UD-IQ4_XS on the 5060 Ti + MTP draft on the 3060 (n_max=2) | llama.cpp master (Oct 2026) | 51.4 | ⚡ spec-decode (MTP) |
| Qwen3.8 27B | i7-12700K + RTX 5060 Ti 16GB + RTX 3060 12GB: Qwen3.8 27B from 21 to 51 tok/s (split, MTP, single-GPU) | mixed / see build | Q4_K_M, split + MTP draft (n_max=2) | llama.cpp master (Oct 2026) | 37.6 | ⚡ spec-decode (MTP) |
| Qwen3.8 27B | i7-12700K + RTX 5060 Ti 16GB + RTX 3060 12GB: Qwen3.8 27B from 21 to 51 tok/s (split, MTP, single-GPU) | mixed / see build | UD-IQ4_XS on the 5060 Ti alone, 8K ctx, q8 KV | llama.cpp master (Oct 2026) | 28.0 | |
| Qwen3.8 27B | i7-12700K + RTX 5060 Ti 16GB + RTX 3060 12GB: Qwen3.8 27B from 21 to 51 tok/s (split, MTP, single-GPU) | mixed / see build | Q4_K_M, layer split 5/3 across both GPUs | llama.cpp (May 2026 build) | 21.2 | |
| Qwen3.8 27B | RTX 3060 12GB: 2026 models (llamabench community) | RTX 3060 12GB | llama-bench tg128 (quant per result) | llama.cpp | 5.2-6.7 | partly in system RAM |
| gpt-oss 20B | Intel Core Ultra 265K + RTX 3060 | RTX 3060 12GB | llama.cpp | 83.96 | partly in system RAM | |
| DeepSeek R1 Distill 14B legacy | Intel Core Ultra 265K + RTX 3060 | RTX 3060 12GB | llama.cpp | 29.77 | ||
| Llama 3.2 3B legacy | Intel Core Ultra 265K + RTX 3060 | RTX 3060 12GB | llama.cpp | 122.85 |
All RTX 3060 12GB builds: see also all GPUs · best LLMs by memory size · download the dataset.
FAQ
How fast is the RTX 3060 12GB for local LLMs?
The fastest plain-decode result we have is 122.85 tok/s on Llama 3.2 3B (, llama.cpp) in Intel Core Ultra 265K + RTX 3060. Speed depends heavily on the model, quant and engine, so compare like with like in the table. RigBase lists 10 measured results from 3 builds.
How much memory does the RTX 3060 12GB have?
12 GB per card. Memory bandwidth: 360 GB/s (memory bandwidth as listed on llamabench).
Which models have been measured on the RTX 3060 12GB?
DeepSeek R1 Distill 14B, Gemma 4 12B, LFM2.5 24B-A2B, Llama 3.2 3B, Qwen3.8 27B, gpt-oss 20B.
Where do these numbers come from?
Every row links to a build page with its original public source (benchmark logs, READMEs, forum posts). RigBase does not re-run the benchmarks. Results marked ⚡ use speculative decoding and are not comparable 1:1 with plain decode.
Referral link: if you sign up with Google and add $10, Runpod gives you and RigBase bonus credit (terms). It never affects our data.