0

DeepSeek R1 Distill Llama 8B

DeepSeek R1 Distill Llama 8B is an AI model from DeepSeek.

Publisher
DeepSeek
Family
Deepseek
Input
Output
Released
20 Jan 2025

Cite

Notes

Only stored in your browser.

Attribution

Benchmark scores
AAOpenLLM
Attribution policy →
Input
$0
/1M tok
Output
$0
/1M tok
Blended
$0
#1/ 500/1M · 3:1
Speed
0
#106/ 499tok/s

Price & speed via Artificial Analysis · ranked across tracked models

Reported on 13 evals across 6 domains - best 85.3% on MATH-500 · #80 of 190

Reported eval scores

13
MATH Level 522
IFBench17.6
AIME 2024: Problems from the American Invitational Mathematics Examination33.3
MATH-50085.3
AIME 2025: Problems from the American Invitational Mathematics Examination41.3
LiveCodeBench23.3
IFEval37.8
Humanity's Last Exam (HLE)4.2
BIG-Bench Hard (BBH)32.4
MMLU-Pro20.9
MuSR32.5
SciCode11.9
GPQA Diamond25.5