Qwen2-7B-Instruct

Qwen

Parameters

Downloads

692.4K

on HF

Likes

687

Avg score

27.9

6 benchmarks

Weights

Open

Released

Jun 4, 2024

About

Qwen2-7B-Instruct is a 7B-parameter open-weights model in the Qwen2 family from Qwen. Benchmarked on 6 evals, averaging 27.9.

apache-2.0text-generation

Benchmark scores · 6

avg 27.9 — higher is better

IFEvalOpen LLM LB · 2026-05

56.8

BIG-Bench HardOpen LLM LB · 2026-05

37.8

MMLU-ProOpen LLM LB · 2026-05

31.6

MATH Lvl 5Open LLM LB · 2026-05

27.6

MuSROpen LLM LB · 2026-05

7.4

GPQA DiamondOpen LLM LB · 2026-05

6.4

The research behind this model

Details

PublisherQwen

FamilyQwen2

Parameters7B

Licenseapache-2.0

WeightsOpen

ReleasedJun 4, 2024

History

Capability over time

Append-only ledger — every observed change to this model's numbers.