Deepthink-Reasoning-7B

prithivMLmods

Parameters

7.6B

Downloads

—

Likes

—

Avg score

29.1

6 benchmarks

Weights

Open

Released

—

About

Deepthink-Reasoning-7B is a 7.6B-parameter open-weights model in the Deepthink family from prithivMLmods. Benchmarked on 6 evals, averaging 29.1.

creativeml-openrail-m

Benchmark scores · 6

avg 29.1 — higher is better

IFEvalOpen LLM LB · 2026-05

48.4

MMLU-ProOpen LLM LB · 2026-05

37.2

BIG-Bench HardOpen LLM LB · 2026-05

35.6

MATH Lvl 5Open LLM LB · 2026-05

33.5

MuSROpen LLM LB · 2026-05

13.4

GPQA DiamondOpen LLM LB · 2026-05

6.6

Details

FamilyDeepthink

Parameters7.6B

Licensecreativeml-openrail-m

WeightsOpen

Released—

History

Capability over time

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