TethysAI_Base_Reasoning

saishshinde15

Parameters

3.1B

Downloads

—

Likes

—

Avg score

26.4

6 benchmarks

Weights

Open

Released

—

About

TethysAI_Base_Reasoning is a 3.1B-parameter open-weights model in the TethysAI family from saishshinde15. Benchmarked on 6 evals, averaging 26.4.

apache-2.0

Benchmark scores · 6

avg 26.4 — higher is better

IFEvalOpen LLM LB · 2026-05

63.7

MATH Lvl 5Open LLM LB · 2026-05

31.4

MMLU-ProOpen LLM LB · 2026-05

24.8

BIG-Bench HardOpen LLM LB · 2026-05

23.6

MuSROpen LLM LB · 2026-05

9.8

GPQA DiamondOpen LLM LB · 2026-05

4.8

Details

FamilyTethysAI

Parameters3.1B

Licenseapache-2.0

WeightsOpen

Released—

History

Capability over time

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