RTX 3060

Nvidia · 12 GB VRAM

19recorded runs

310.4best tok/s

13sourced

How it compares — 12 GB tier

RTX 3060310.4 RTX 5070185.8

Best verified tok/s per GPU in this VRAM tier — compare any two like-for-like.

Evidence: 13 sourced runs · 6 awaiting review · 2 independent sources · first recorded Jul 2, 2026 · last recorded Sep 5, 2026. What these labels mean.

What fits on 12 GB — measured

Model	Quant	Peak VRAM	Headroom
gemma-4-12B-it	Q4_K_XL	9.3 GB	2.7 GB
Ling-3.0-tiny	Q4_K_M	8.0 GB	4.0 GB
gemma-4-12B-it	Q4_K_M	7.6 GB	4.4 GB
granite-4.2-8b	Q5_K_M	6.6 GB	5.4 GB
Qwythos-9B-v2	Q4_K_M	5.8 GB	6.2 GB
Qwen3.5-9B	Q4_K_M	5.7 GB	6.3 GB
Ornith-1.0-9B	Q4_K_M	5.7 GB	6.3 GB
Qwen3.5-4B	Q4_K_M	3.5 GB	8.5 GB

Peak process memory recorded by the submitter on this card — evidence of what fit, not a guarantee. Plan with the VRAM calculator.

Price history

new$495▲ 3.8%3 snapshots

Weekly street-price snapshots — all snapshots.

All benchmarks Compare RTX 3060 →

GPU	Model	Quant	Backend	Ctx	tok/s	prompt tok/s	VRAM GB	Source
RTX 3060 · 12 GB	Qwen3-0.6B	Q8_0	llama.cpp b9775 / 7529fdaaf	4608	310.4 ✓	13631.6	2.8	src
RTX 3060 · 12 GB	Spark-X2.5-1.7B	IQ4_XS	llama.cpp a698f1cc (XHToken spark2_5 fork b105…	4096	207.3 ✓	7951.0	3.0	src
RTX 3060 · 12 GB	gemma-4-E2B-it	Q4_K_M	llama.cpp b10443	32768	113.4 ✓	3555.4	1.9	src
RTX 3060 · 12 GB	gemma-4-12B-it	Q4_K_XL	llama.cpp b9775 / be4a6a63e	4096	89.7 ✓	114.9	9.3	src
RTX 3060 · 12 GB	Spark-X2.5-4B	Q4_K_M	llama.cpp a698f1cc (XHToken spark2_5 fork b105…	4096	84.7 ✓	3346.7	3.4	src
RTX 3060 · 12 GB	Qwen3.5-4B	Q4_K_M	llama.cpp b10443	32768	78.6 ✓	2111.9	3.5	src
RTX 3060 · 12 GB	gemma-4-E4B-it	Q4_K_M	llama.cpp b10443	32768	70.9 ✓	2051.0	3.5	src
RTX 3060 · 12 GB	Ling-3.0-tiny	Q4_K_M	llama.cpp d8d8625 (PR #26608 bailingmoe3, unca…	262144	52.6 ✓	730.6	8.0	src
RTX 3060 · 12 GB	granite-4.2-8b	Q5_K_M	llama.cpp 0.1.2-dev (build 197, commit b062ba7…	4096	52.5 ✓	1708.0	6.6	src
RTX 3060 · 12 GB	Ornith-1.0-9B	Q4_K_M	llama.cpp b10443	32768	52.5 ✓	1457.6	5.7	src
RTX 3060 · 12 GB	Qwen3.5-9B	Q4_K_M	llama.cpp b10443	32768	51.9 ✓	1457.9	5.7	src
RTX 3060 · 12 GB	Qwythos-9B-v2	Q4_K_M	llama.cpp b10443	32768	51.1 ✓	1446.7	5.8	src
RTX 3060 · 12 GB	gemma-4-12B-it	Q4_K_M	llama.cpp b10443	32768	37.0 ✓	892.7	7.6	src
RTX 3060 · 12 GB	Huihui-Qwythos-9B-Claude-Mythos-5-1M-abliterated-i1	IQ4_XS	llama.cpp —	65536	90.2 …	611.2	—	src
RTX 3060 · 12 GB	Qwen3.8-9B-Distill	Q4_K_M	llama.cpp —	131072	76.4 …	560.7	—	src
RTX 3060 · 12 GB	Ornith-1.5-9B	Q4_K_M	llama.cpp —	65536	65.0 …	3114.4	—	src
RTX 3060 · 12 GB	Ornith-1.5-35B-A3B	APEX-MTP-I-Compact	llama.cpp —	65536	54.3 …	1341.4	—	src
RTX 3060 · 12 GB	Qwen3.8-27B	IQ2_XS-MTP	llama.cpp GSQ-RCO+Unsloth MTP	82000	38.0 …	460.0	—	src
RTX 3060 · 12 GB	Qwen3.8-27B	IQ2_XS	llama.cpp GSQ-RCO	158000	22.0 …	460.0	—	src

Related reading: GPU buying guide · used GPUs · Value index · Street prices