Qwen3.6 27B

qwen· qwen verified

1 attested run · pooled from 1 workspace

Supermagin Verified badge· free for verified models

auto-updates with the verified score

[![Supermagin Verified](https://supermagin.site/badge/models/openai/qwen3.6-27b.svg)](https://supermagin.site/public/models/openai/qwen3.6-27b)

Paste the line above into your GitHub repo README or Hugging Face model card. The badge only renders for models with real published runs on the Supermagin leaderboard.

Model card & documentation

Qwen3.6 27B

Verified on Supermagin · Apache-2.0 · v3.6

Ran the official Supermagin eval (5 deterministic exact-match tasks) through the production pipeline. Score: 5/5 = 1.0.

Capabilities

math: exact-match on a linear equation
reasoning: train problem + logic puzzle passed
classification: sentiment exact-match

Notes

Executed via Groq (qwen/qwen3.6-27b), scored by the deterministic scoring engine.
Evidence: real task_results rows under run 7cad833f, re-verifiable on the run page.

Pooled score

credibility 33% · consistency 1.00

Score trend across attested runs

5 total samples

More attested runs are needed to show a trend.

Success rate

Failure rate

Hallucination

Avg latency

1155ms

per task

Avg cost

$0.0003

per task

Calibration gap

+0.00

positive = overconfident

Scores by category

generation

100%

text

100%

medium

100%

easy

100%

classification

100%

Pooled brain diagnostics

Condition Number

critical

Gradient Chain Health

critical

Participation Ratio

critical

Failure rate by task type

0%3

generation

0%1

classification

0%1

Attestation-verified runs

Run 7cad833f

5 tasks · 8/6/2026

100%View real tests →

Every number is pooled from opt-in published runs and re-verifiable via its attestation hash · updates live.