Benchmark Almanac

Yi-1.5-34B

Yi-1.5-34B.

01.AI Released May 13th, 2024 Compare with others
21
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
01.AI May 2024

Yi-1.5-34B

Open weights

No Epoch Capabilities Index score yet
  • GPQA Diamond 32% #205 /229
  • KOR-Bench 39.8% #20 /33
  • BigCodeBench-Hard 9.20% #85 /96

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Knowledge and language

BenchmarkScoreRankSource
WildBench 47.4 #16 of 47 Official leaderboard
Vectara Hallucination Leaderboard (HHEM) 3.70% #55 of 148 Official leaderboard
MuSR 42.4% #39 of 98 Official leaderboard
MMLU-CF 61.3% #16 of 37 Official leaderboard
MixEval 81.7% #16 of 34 Official leaderboard
EQ-Bench v2 72.9 #45 of 85 Official leaderboard
MMLU-Pro 52.3% #130 of 212 Official leaderboard
Chinese SimpleQA 31.8% #17 of 27 Official leaderboard
Open Arabic LLM Leaderboard v1 38.9% #41 of 55 Official leaderboard
SuperGPQA 26% #46 of 59 Official leaderboard
LMArena (Elo) 1,213 #279 of 335 Official leaderboard

Math and science

BenchmarkScoreRankSource
MathBench 52.2% #9 of 21 Official leaderboard
PHYSICS 17.4% #10 of 20 Official leaderboard
KOR-Bench 39.8% #20 of 33 Official leaderboard
MATH Level 5 25.5% #69 of 94 Epoch AI run
ZebraLogic 11.5% #35 of 46 Official leaderboard
GPQA Diamond 32% #205 of 229 Epoch AI run

Coding

BenchmarkScoreRankSource
BigCodeBench 38.8% #48 of 69 Official leaderboard
BigCodeBench-Hard 9.20% #85 of 96 Official leaderboard

Vision, audio and robotics

BenchmarkScoreRankSource
SGP-Bench 44.3% #16 of 22 Official leaderboard

Other

BenchmarkScoreRankSource
Open LLM Leaderboard v2 25.6 #41 of 99 Official leaderboard