Meta-Llama-3-8B-Instruct

Llama family · 8.0B params

1recorded run

148.3best tok/s

0verified

Evidence: 0 sourced runs · 1 awaiting review · 1 independent source · first recorded Jun 29, 2026. What these labels mean.

What the Meta-Llama-3-8B-Instruct numbers mean

The fastest result on our board is 148.3 output tok/s on an RTX 5080 at Q4_0 with a 4096-token context. That is about 6.7 ms per generated token — 500 tokens takes roughly 3.4 seconds. Prompt processing on the same record runs at 8429.6 tok/s, so a 2,000-token prompt is read in about 0.2 seconds.

How much VRAM does Meta-Llama-3-8B-Instruct need?

Meta-Llama-3-8B-Instruct has 8.0B parameters. These estimates use the same formula as the VRAM calculator: weights, plus about 15% for a moderate-context KV cache and runtime overhead, plus a small engine floor.

Quantisation	Estimated VRAM	Smallest card that fits
Q4_K_M (4-bit)	6.0 GB	8 GB
Q8_0 (8-bit)	10.6 GB	16 GB
FP16 (16-bit)	19.2 GB	24 GB

Long contexts and large batches raise the real number — read the smallest card as "possible with a reduced context". Measured throughput is on the benchmark board.

Fastest GPUs for this model (best tok/s)

RTX 5080148.3

Models with a similar footprint

Model	Params	Q4_K_M weights
AceReason-Nemotron-1.1-7B	8B	4.5 GB
DeepSeek-R1-0528-Qwen3-8B	8B	4.5 GB
DeepSeek-R1-Distill-Qwen-7B	8B	4.5 GB
gemma-4-E4B-it	8B	4.5 GB
granite-4.2-8b	8B	4.5 GB
InternVL3-8B	8B	4.5 GB

Ranked by distance from Meta-Llama-3-8B-Instruct's parameter count. Full benchmark board.

All benchmarks Methodology

GPU	Model	Quant	Backend	Ctx	tok/s	prompt tok/s	VRAM GB	Source
RTX 5080 · 16 GB	Meta-Llama-3-8B-Instruct	Q4_0	llama.cpp —	4096	148.3 …	8429.6	—	src