Hardware

NVIDIA A100 80GB PCIe

NVIDIA A100 80GB PCIe has 80 GB HBM2e and 1935 GB/s. See which open models fit, at which quant, and roughly how fast. Figures below use the default memory (80 GB), 32 GB system RAM, and DDR5-5600 for anything that spills. Updated 2026-10-17.

Memory80 GB HBM2e

Bandwidth1935 GB/s

Kindgpu

Year2021

PCIe 80GB, 1935 GB/s. Source.

RSS for new models on this device

Models that fit

Model	Verdict	Quant	Context	tok/s
Mistral Medium 3.5 128B 125B	Tight fit	UD-Q4_K_XL	8K	14–18
Devstral 2 123B Instruct 2512 125B	Tight fit	UD-Q4_K_XL	8K	14–18
gpt oss 120b 117B	Runs	UD-Q8_K_XL	8K	87–118
Qwen3.5 122B A10B MTP 125B	Tight fit	UD-Q4_K_XL	4K	166–226
Mistral Small 4 119B 2603 119B	Tight fit	UD-Q4_K_XL	8K	78–106
Llama 4 Scout 17B 16E Instruct 108B	Tight fit	Q5_K_M	8K	60–81
GLM 4.6V 107B	Tight fit	Q5_K_S	8K	60–82
GLM 4.5 Air 110B	Tight fit	UD-Q4_K_XL	8K	67–91
Qwen2.5 72B Instruct 72.7B	Tight fit	Q8_0	4K	14–18
Llama 3.3 70B Instruct 70.6B	Tight fit	Q8_0	8K	14–19
DeepSeek R1 Distill Llama 70B 70.6B	Tight fit	Q8_0	8K	14–19
Hermes 4 70B 70.6B	Tight fit	Q8_0	8K	14–19
NVIDIA Nemotron 3 Super 120B A12B 121B	Runs	IQ4_NL	8K	149–203
Qwen3 Coder Next 79.7B	Tight fit	UD-Q6_K_XL	8K	81–110
Qwen3 Next 80B A3B Instruct 79.7B	Tight fit	UD-Q6_K_XL	8K	314–429
Qwen3 Next 80B A3B Thinking 79.7B	Tight fit	UD-Q6_K_XL	8K	314–429
Step 3.7 Flash 197B	Tight fit	UD-Q2_K_XL	8K	45–61
MiniMax M2 229B	Tight fit	UD-IQ2_XXS	8K	69–94
Qwen3.8 Flash Next 177B	Tight fit	UD-IQ1_M	8K	83–113
Qwen2.5 Coder 32B Instruct 32.8B	Runs great	Q8_0	8K	29–39
Qwen2.5 32B Instruct 32.8B	Runs	F16	8K	16–21
Qwen3 32B 32.8B	Runs	BF16	8K	16–21
DeepSeek R1 Distill Qwen 32B 32.8B	Runs	F16	8K	16–21
Qwen3 VL 32B Instruct 32.8B	Runs	BF16	8K	16–21
GLM 4 32B 0414 32.6B	Runs	BF16	8K	16–22
gemma 4 31B it 30.7B	Runs	BF16	8K	15–21
Qwen3.8 27B 27.3B	Runs	BF16	8K	19–26
gemma 3 27b it 27.0B	Runs	BF16	8K	18–25
medgemma 27b text it 27.0B	Runs	BF16	8K	18–25
Qwen3.6 27B 26.9B	Runs	BF16	8K	20–27
Qwen3.5 27B 26.9B	Runs	BF16	8K	20–27
Kwaipilot KAT Coder V2.5 Dev 34.7B	Tight fit	BF16	8K	84–114
Qwen3.8 35B A3B Distill 35.5B	Tight fit	BF16	8K	172–235
Qwen3.6 35B A3B 34.7B	Tight fit	BF16	8K	172–235
Qwen3.5 35B A3B 34.7B	Tight fit	BF16	8K	172–235
NVIDIA Nemotron 3.5 Lightning 30B A3B 32.9B	Runs	BF16	8K	165–225
Devstral Small 2 24B Instruct 2512 23.6B	Runs	BF16	8K	22–30
Mistral Small 3.2 24B Instruct 2506 23.6B	Runs	BF16	8K	22–30
Magistral Small 2509 23.6B	Runs	BF16	8K	22–30
NVIDIA Nemotron 3 Nano Omni 30B A3B Reasoning 31.6B	Runs great	UD-Q8_K_XL	8K	280–382

90 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 NVIDIA A100 80GB PCIe at UD-Q4_K_XL, 8K context, about 14–18 tok/s. Weights 70.5 GB plus 2.75 GB of KV cache — 74.2 GB of 76.0 GB usable. Weights shown in the verdict use GiB (the same 1024³ counting as GPU memory).