NVIDIA · Blackwell

What runs on a RTX 5080

16 GB at 960 GB/s. The largest model that fits at Q4_K_M is gpt-oss 20B-A3.6B, at roughly 195 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 14.2 GB · 45%

014 GB

Weights 4.6 GB

KV cache 1.0 GB

Runtime overhead 0.8 GB

Llama 3.1 8B at Q4_K_M leaves 7.8 GB spare on a RTX 5080. There is room to raise the context length or move up a quantisation level.

Generation121tok/s

Prompt processing2929tok/s

Max context70Ktokens

KV per 1K tokens0GB

Every quantisation of Llama 3.1 8B on a RTX 5080

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	Won't fit	—	11.3
INT8 / W8A8	8.50	7.7 GB	9.5 GB	Runs comfortably	46K	76.2
Q8_0 (GGUF)	8.50	7.7 GB	9.5 GB	Runs comfortably	46K	76.2
FP8 (E4M3)	8.00	7.5 GB	9.3 GB	Runs comfortably	48K	78.3
Q6_K	6.56	6.1 GB	8.0 GB	Runs comfortably	58K	93.9
AWQ 4-bit	4.25	5.4 GB	7.2 GB	Runs comfortably	64K	105
GPTQ 4-bit	4.25	5.4 GB	7.2 GB	Runs comfortably	64K	105
MXFP4	4.25	5.4 GB	7.2 GB	Runs comfortably	64K	105
Q5_K_M	5.67	5.3 GB	7.1 GB	Runs comfortably	65K	107
Q5_K_S	5.52	5.2 GB	7.0 GB	Runs comfortably	66K	110
Q4_K_M	4.85	4.6 GB	6.4 GB	Runs comfortably	70K	121
Q4_K_S	4.58	4.4 GB	6.2 GB	Runs comfortably	72K	127
Q4_0	4.55	4.4 GB	6.2 GB	Runs comfortably	72K	127
IQ4_XS	4.25	4.1 GB	5.9 GB	Runs comfortably	74K	134
Q3_K_M	3.91	3.8 GB	5.7 GB	Runs comfortably	77K	142
IQ3_M	3.70	3.7 GB	5.5 GB	Runs comfortably	78K	148
IQ3_XXS	3.06	3.2 GB	5.0 GB	Runs comfortably	82K	168
Q2_K	2.63	2.8 GB	4.6 GB	Runs comfortably	85K	185
IQ2_XXS	2.06	2.3 GB	4.2 GB	Runs comfortably	89K	214
IQ1_M	1.75	2.1 GB	3.9 GB	Runs comfortably	91K	234

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	—	2.03
Qwen3 235B-A22B	235.1B	132.8 GB	Won't fit	—	3.36
gpt-oss 120B-A5.1B	116.8B	66.0 GB	Won't fit	—	15.9
Llama 4 Scout 109B-A17B	109B	61.7 GB	Won't fit	—	4.91
Qwen2.5 72B	72.7B	41.2 GB	Won't fit	—	1.31
Llama 3.1 70B	70.5B	40.0 GB	Won't fit	—	1.35
Mixtral 8x7B	46.7B	26.4 GB	Won't fit	—	9.48
Command R 35B	35.0B	20.1 GB	Won't fit	—	1.96
Yi-1.5 34B	34.4B	19.5 GB	Won't fit	—	4.53
Qwen3 32B	32.8B	18.6 GB	Won't fit	—	4.98
Qwen2.5-Coder 32B	32.8B	18.6 GB	Won't fit	—	4.98
DeepSeek-R1-Distill-Qwen 32B	32.8B	18.6 GB	Won't fit	—	4.98
Qwen3 30B-A3B	30.5B	17.3 GB	Won't fit	—	46.2
Gemma 3 27B	27.4B	15.6 GB	Won't fit	—	9.29
Mistral Small 24B	23.6B	13.4 GB	Won't fit	—	18.7
gpt-oss 20B-A3.6B	20.9B	11.9 GB	Fits, but tight	33K	195
Qwen3 14B	14.8B	8.5 GB	Runs comfortably	32K	68.7
Phi-4 14B	14.7B	8.4 GB	Runs comfortably	16K	68.1
Mistral NeMo 12B	12.3B	7.0 GB	Runs comfortably	41K	81.4
Gemma 3 12B	12.2B	7.0 GB	Runs comfortably	98K	82.3
GLM-4 9B	9.4B	5.4 GB	Runs comfortably	128K	111
Qwen3 8B	8.2B	4.7 GB	Runs comfortably	62K	117
Llama 3.1 8B	8.0B	4.6 GB	Runs comfortably	70K	121
DeepSeek-R1-Distill-Llama 8B	8.0B	4.6 GB	Runs comfortably	70K	121
Qwen2.5 7B	7.6B	4.4 GB	Runs comfortably	128K	135
Mistral 7B v0.3	7.3B	4.1 GB	Runs comfortably	32K	134
Gemma 3 4B	4.3B	2.5 GB	Runs comfortably	128K	226
Qwen3 4B	4.0B	2.3 GB	Runs comfortably	79K	212
Llama 3.2 3B	3.2B	1.8 GB	Runs comfortably	106K	266
Llama 3.2 1B	1.2B	0.7 GB	Runs comfortably	128K	700