Leaderboard

EvalPlus

evalplus

Record

Category
coding
Scored models
9

Scores

ModelScore
JetBrains/Mellum2-12B-A2.5B-Instruct76.2
tencent/Hunyuan-A13B-Instruct75.69
Qwen/Qwen2-72B75.2
Qwen/Qwen2-7B60.9
Qwen/Qwen2-72B55
Qwen/Qwen2-57B-A14B48.5
Qwen/Qwen2-57B-A14B46.4
tencent/Hunyuan-7B-Pretrain39.71
Qwen/Qwen2-7B36.4

Remaining fields

Identity, launch, related records, and measured speed are shown above. This is the rest of the normalized record.

category
coding
id
evalplus
name
EvalPlus
schema version
local-ai-registry/v1

rows

variantconfcontextmodel idorgrankrootscore
Mellum2-12B-A2.5B-Instruct-GGUFmediumEvalPlusmellum2-12b-a2-5b-instructMungert1JetBrains/Mellum2-12B-A2.5B-Instruct76.2
Hunyuan-A13B-Instruct-GGUFmediumEvalPlushunyuan-a13b-instructunsloth2tencent/Hunyuan-A13B-Instruct75.69
Qwen2-72B-InstructmediumEvalPlusqwen2-72bQwen3Qwen/Qwen2-72B75.2
Qwen2-7B-InstructmediumEvalplusqwen2-7bQwen4Qwen/Qwen2-7B60.9
Qwen2-72BmediumEvalPlusqwen2-72bQwen5Qwen/Qwen2-72B55
Qwen2-57B-A14B-InstructmediumEvalPlusqwen2-57b-a14bQwen6Qwen/Qwen2-57B-A14B48.5
Qwen2-57B-A14BmediumEvalPlusqwen2-57b-a14bQwen7Qwen/Qwen2-57B-A14B46.4
Hunyuan-7B-Instruct-GGUFmediumEvalPlushunyuan-7b-pretrainMungert8tencent/Hunyuan-7B-Pretrain39.71
Qwen2-7BmediumEvalPlusqwen2-7bQwen9Qwen/Qwen2-7B36.4
Provenance & metadata (1)

source

kind
leaderboard-scrape
paths
benchmarks/evalplus.html