Qwen2.5 7B Instruct

Open weights

Qwen2.5-7B-Instruct is an AI model with 7.0B parameters, released with open weights.

Publisher: Alibaba
Family: Qwen
Params: 7.0B
Context: 131K
Input
Output
License: unknown
Released: 16 Sep 2024
Canonical: huggingface.co/Qwen/Qwen2.5-7B-Instruct

Cite

Notes

Only stored in your browser.

Attribution

Model card: huggingface.co/Qwen/Qwen2.5-7B-Instruct/raw/main/README.md
Benchmark scores: OpenLLM prime-hub

Attribution policy →

Context

131K

Reported on 8 evals across 4 domains - best 75.9% on IFEval · #13 of 81

Reported eval scores

8

BIG-Bench Hard (BBH)53.9

MuSR40.2

Blackjack Env-0.1

GPQA Diamond29.1