R1 Distill Llama 70B

Llama / DeepSeek

28scores
24benchmarks
$0.7 / $0.8 per 1M tokenscost in/out

Metadata

Llama Open source

Aliases: deepseek-deepseek-r1-distill-llama-70b, deepseek-r1-distill-llama-70b, deepseek/deepseek-r1-distill-llama-70b

Benchmark Results

Benchmark Category Rank Score Sampled
AgentIF Agentic 11 55 2026-05-27
Tau2-Bench Telecom Agentic 303 21.9% 2026-05-11
Terminal-Bench Hard Agentic 327 1.5% 2026-05-11
BigCodeBench Coding 86 35.30 2026-05-06
SciCode Coding 235 31.2% 2026-05-11
AI Energy Score Efficiency 90 5 2026-05-06
AI Energy Score Efficiency 200 1 2026-05-06
Open LLM Leaderboard v2 General Knowledge 1320 27.81 2026-05-06
HealthBench Hard Healthcare 13 0.47 2026-05-27
Artificial Analysis Intelligence Index Intelligence 304 15.95 2026-05-11
Humanity's Last Exam Intelligence 238 6.1% 2026-05-11
MMLU-Pro Intelligence 123 79.5% 2026-05-11
MuSR Intelligence 1218 13.28 2026-05-06
Open Japanese LLM Leaderboard Language 236 58.27 2026-05-06
Open Japanese LLM Leaderboard Language 754 22.80 2026-05-06
Open Portuguese LLM Leaderboard Language 760 59.63 2026-05-06
AIME 2025 Math 133 53.7% 2026-05-11
MATH Level 5 Math 801 30.74 2026-05-06
MATH-500 Mathematics 20 0.94 2026-05-06
BRIDGE Medical Leaderboard Medical 41 46.17 2026-05-27
BRIDGE Medical Leaderboard Medical 106 39.79 2026-05-27
BRIDGE Medical Leaderboard Medical 121 38.95 2026-05-27
MEDIC Benchmark Medical 27 68.84 average normalized public table score 2026-05-27
Artificial Analysis Openness Index Openness 157 36.11 2026-05-11
GPQA Diamond Reasoning 396 40.2% 2026-05-11
CritPt Science 168 0% 2026-05-11
Defects4J Software Engineering 31 0.221 2026-05-27
RepairBench Software Engineering 30 0.208 2026-05-27