Hardware

GeForce RTX 3080 10GB

GeForce RTX 3080 10GB has 10 GB GDDR6X and 760 GB/s. See which open models fit, at which quant, and roughly how fast. Figures below use the default memory (10 GB), 32 GB system RAM, and DDR5-5600 for anything that spills. Updated 2026-10-24.

Memory10 GB GDDR6X

Bandwidth760 GB/s

Kindgpu

Year2020

320-bit GDDR6X at 19 Gbps = 760 GB/s. Source.

RSS for new models on this device

Models that fit

Model	Verdict	Quant	Context	tok/s
Ministral 3 14B Instruct 2512 13.5B	Tight fit	UD-Q4_K_XL	4K	46–63
Ministral 3 14B Reasoning 2512 13.5B	Tight fit	UD-Q4_K_XL	4K	46–63
gemma 4 12b it 11.9B	Tight fit	Q5_K_M	4K	42–57
Qwen3 14B 14.8B	Tight fit	IQ4_XS	4K	47–65
NousResearch Hermes 4 14B 14.8B	Tight fit	IQ4_XS	4K	48–65
Qwen2.5 14B Instruct 14.8B	Tight fit	IQ4_XS	4K	47–64
Qwen3 32B 32.8B	Tight fit	UD-IQ1_S	4K	47–65
Qwen3 VL 32B Instruct 32.8B	Tight fit	UD-IQ1_S	4K	47–65
GLM 4 32B 0414 32.6B	Tight fit	UD-IQ1_M	8K	48–65
GLM 4.6V Flash 9.4B	Tight fit	UD-Q6_K_XL	4K	46–63
GLM Z1 9B 0414 9.4B	Tight fit	UD-Q6_K_XL	4K	46–63
GLM 4 9B 0414 9.4B	Tight fit	UD-Q6_K_XL	4K	46–63
Qwen3.6 14B A3B FableVibes 13.8B	Tight fit	Q4_K_M	8K	208–283
Qwen3.8 9B Distill 9.2B	Tight fit	Q6_K	8K	53–73
MiMo V2.6 Distill Qwen 9B 9.0B	Runs	Q5_K_M	8K	59–80
Qwythos 9B Claude Mythos 5 1M 9.0B	Runs	Q6_K	8K	55–75
Qwythos 9B v2 9.0B	Runs	Q6_K	8K	54–74
Qwen3.5 9B 9.0B	Tight fit	UD-Q6_K_XL	8K	46–63
granite 4.2 8b 8.8B	Tight fit	Q6_K	8K	49–67
Ministral 3 8B Instruct 2512 8.5B	Tight fit	UD-Q6_K_XL	8K	47–64
Ministral 3 8B Reasoning 2512 8.5B	Tight fit	UD-Q6_K_XL	8K	47–64
DeepSeek R1 Distill Qwen 14B 14.8B	Tight fit	Q3_K_M	8K	47–64
Qwen2.5 Coder 14B Instruct 128K 14.8B	Tight fit	Q3_K_M	8K	47–64
Qwen3.8 27B 27.3B	Tight fit	UD-IQ2_S	8K	47–64
phi 4 14.7B	Tight fit	Q3_K_M	8K	47–64
Qwen3 VL 8B Instruct 8.2B	Tight fit	UD-Q6_K_XL	8K	48–66
DeepSeek R1 0528 Qwen3 8B 8.2B	Tight fit	UD-Q6_K_XL	8K	48–66
Qwen3 8B 8.2B	Tight fit	UD-Q6_K_XL	8K	48–66
Qwen3 VL 8B Thinking 8.2B	Tight fit	UD-Q6_K_XL	8K	48–66
Qwen3.5 27B 26.9B	Tight fit	UD-IQ2_XXS	4K	47–64
gemma 3 27b it 27.0B	Tight fit	UD-IQ2_XXS	8K	35–47
medgemma 27b text it 27.0B	Tight fit	UD-IQ2_XXS	8K	35–47
Llama 3.1 8B Instruct 8.0B	Tight fit	UD-Q6_K_XL	8K	50–68
Apertus 8B Instruct 2509 8.1B	Tight fit	UD-Q6_K_XL	8K	49–67
DeepSeek R1 Distill Llama 8B 8.0B	Tight fit	Q8_0	4K	46–63
Qwen2.5 VL 7B Instruct 7.6B	Tight fit	Q8_0	8K	49–67
Qwen2.5 7B Instruct 7.6B	Tight fit	Q8_0	8K	49–67
DeepSeek R1 Distill Qwen 7B 7.6B	Tight fit	Q8_0	8K	49–67
Qwen2.5 Coder 7B Instruct 128K 7.6B	Tight fit	Q8_0	8K	49–67
NVIDIA Nemotron 3.5 Bolt E4B 7.5B	Tight fit	UD-Q8_K_XL	8K	47–64

105 more models in the catalog do not fit this default profile, or ranked below the first 40 that do.

Example: Tight fit fully on your GeForce RTX 3080 10GB at UD-Q4_K_XL, 4K context, about 46–63 tok/s. Weights 7.79 GB plus 0.63 GB of KV cache — 8.79 GB of 9.03 GB usable. Weights shown in the verdict use GiB (the same 1024³ counting as GPU memory).