NVIDIA · Ampere

What runs on a A100 80GB

80 GB at 2039 GB/s. The largest model that fits at Q4_K_M is gpt-oss 120B-A5.1B, at roughly 227 tokens per second.

Model

Hardware

Quantisation

Context 8K

KV cache precision

Runtime

Batch

GPUs

Monitor plugged into this GPU

Runs comfortably

6.4 GB of 74.4 GB · 9%

074 GB

Weights 4.6 GB

KV cache 1.0 GB

Runtime overhead 0.8 GB

Llama 3.1 8B at Q4_K_M leaves 68.0 GB spare on a A100 80GB. There is room to raise the context length or move up a quantisation level.

Generation238tok/s

Prompt processing8159tok/s

Max context128Ktokens

KV per 1K tokens0GB

Every quantisation of Llama 3.1 8B on a A100 80GB

Highlighted row is the highest quality that still fits at 8K context.

Quantisation	bpw	Weights	Total	Verdict	Max ctx	tok/s
FP16 / BF16	16.00	15.0 GB	16.8 GB	Runs comfortably	128K	81.3
INT8 / W8A8	8.50	7.7 GB	9.5 GB	Runs comfortably	128K	151
Q8_0 (GGUF)	8.50	7.7 GB	9.5 GB	Runs comfortably	128K	151
FP8 (E4M3)	8.00	7.5 GB	9.3 GB	Runs comfortably	128K	155
Q6_K	6.56	6.1 GB	8.0 GB	Runs comfortably	128K	186
AWQ 4-bit	4.25	5.4 GB	7.2 GB	Runs comfortably	128K	207
GPTQ 4-bit	4.25	5.4 GB	7.2 GB	Runs comfortably	128K	207
MXFP4	4.25	5.4 GB	7.2 GB	Runs comfortably	128K	207
Q5_K_M	5.67	5.3 GB	7.1 GB	Runs comfortably	128K	211
Q5_K_S	5.52	5.2 GB	7.0 GB	Runs comfortably	128K	216
Q4_K_M	4.85	4.6 GB	6.4 GB	Runs comfortably	128K	238
Q4_K_S	4.58	4.4 GB	6.2 GB	Runs comfortably	128K	248
Q4_0	4.55	4.4 GB	6.2 GB	Runs comfortably	128K	250
IQ4_XS	4.25	4.1 GB	5.9 GB	Runs comfortably	128K	262
Q3_K_M	3.91	3.8 GB	5.7 GB	Runs comfortably	128K	278
IQ3_M	3.70	3.7 GB	5.5 GB	Runs comfortably	128K	288
IQ3_XXS	3.06	3.2 GB	5.0 GB	Runs comfortably	128K	327
Q2_K	2.63	2.8 GB	4.6 GB	Runs comfortably	128K	359
IQ2_XXS	2.06	2.3 GB	4.2 GB	Runs comfortably	128K	413
IQ1_M	1.75	2.1 GB	3.9 GB	Runs comfortably	128K	449

Q4_K_M at 8K context. Bandwidth sets the speed; capacity sets the ceiling.

Model	Params	Weights	Verdict	Max ctx	tok/s
DeepSeek-R1 671B-A37B	671B	379.0 GB	Won't fit	—	2.26
Qwen3 235B-A22B	235.1B	132.8 GB	Won't fit	—	6.02
gpt-oss 120B-A5.1B	116.8B	66.0 GB	Fits, but tight	108K	227
Llama 4 Scout 109B-A17B	109B	61.7 GB	Runs comfortably	63K	100
Qwen2.5 72B	72.7B	41.2 GB	Runs comfortably	103K	29.6
Llama 3.1 70B	70.5B	40.0 GB	Runs comfortably	107K	30.5
Mixtral 8x7B	46.7B	26.4 GB	Runs comfortably	32K	129
Command R 35B	35.0B	20.1 GB	Runs comfortably	43K	50.3
Yi-1.5 34B	34.4B	19.5 GB	Runs comfortably	32K	61.1
Qwen3 32B	32.8B	18.6 GB	Runs comfortably	128K	63.5
Qwen2.5-Coder 32B	32.8B	18.6 GB	Runs comfortably	128K	63.5
DeepSeek-R1-Distill-Qwen 32B	32.8B	18.6 GB	Runs comfortably	128K	63.5
Qwen3 30B-A3B	30.5B	17.3 GB	Runs comfortably	128K	163
Gemma 3 27B	27.4B	15.6 GB	Runs comfortably	128K	76.0
Mistral Small 24B	23.6B	13.4 GB	Runs comfortably	32K	89.0
gpt-oss 20B-A3.6B	20.9B	11.9 GB	Runs comfortably	128K	327
Qwen3 14B	14.8B	8.5 GB	Runs comfortably	128K	136
Phi-4 14B	14.7B	8.4 GB	Runs comfortably	16K	135
Mistral NeMo 12B	12.3B	7.0 GB	Runs comfortably	128K	161
Gemma 3 12B	12.2B	7.0 GB	Runs comfortably	128K	162
GLM-4 9B	9.4B	5.4 GB	Runs comfortably	128K	218
Qwen3 8B	8.2B	4.7 GB	Runs comfortably	128K	230
Llama 3.1 8B	8.0B	4.6 GB	Runs comfortably	128K	238
DeepSeek-R1-Distill-Llama 8B	8.0B	4.6 GB	Runs comfortably	128K	238
Qwen2.5 7B	7.6B	4.4 GB	Runs comfortably	128K	265
Mistral 7B v0.3	7.3B	4.1 GB	Runs comfortably	32K	263
Gemma 3 4B	4.3B	2.5 GB	Runs comfortably	128K	434
Qwen3 4B	4.0B	2.3 GB	Runs comfortably	128K	407
Llama 3.2 3B	3.2B	1.8 GB	Runs comfortably	128K	512
Llama 3.2 1B	1.2B	0.7 GB	Runs comfortably	128K	1319