HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

Home/Models/nvidia/Llama-3.1-8B-Medusa-FP8

Llama-3.1-8B-Medusa-FP8

nvidia/Llama-3.1-8B-Medusa-FP8

kind: fp8
quant: fp8 (8-bit)
family: llama
modality: —
root model: —
provider: nvidia
license: —
language: —
context: 128K
library: —
vRAM (GB): —
size on disk: —
downloads: 50
likes: 12
created: —
last modified: 2025-08-22T21:25:28+00:00 (2025)
huggingface: https://huggingface.co/nvidia/Llama-3.1-8B-Medusa-FP8
tags: safetensorsllamaregion:us

Benchmark scores (2)

benchmark	category	score	context	conf	evidence
MMLU	knowledge	68.3		high	`\| FP8 \| 68.3 \| 2.07 \|`
MT-Bench	chat	2.07		high	`\| FP8 \| 68.3 \| 2.07 \|`

Discussions (0)

none scraped.

Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.