Qwen2-VL-7B-Instruct

Qwen

Parameters

8.3B

Downloads

—

Likes

—

Avg score

26.5

6 benchmarks

Weights

Open

Released

—

About

Qwen2-VL-7B-Instruct is a 8.3B-parameter open-weights model in the Qwen2 family from Qwen. Benchmarked on 6 evals, averaging 26.5.

apache-2.0

Benchmark scores · 6

avg 26.5 — higher is better

IFEvalOpen LLM LB · 2026-05

46.0

BIG-Bench HardOpen LLM LB · 2026-05

35.9

MMLU-ProOpen LLM LB · 2026-05

34.4

MATH Lvl 5Open LLM LB · 2026-05

19.9

MuSROpen LLM LB · 2026-05

13.6

GPQA DiamondOpen LLM LB · 2026-05

9.3

Details

PublisherQwen

FamilyQwen2

Parameters8.3B

Licenseapache-2.0

WeightsOpen

Released—

History

Capability over time

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