Kingy AI
Work with Kingy

Kingy evidence library

Reviews & Tests

See the results. Check the limits. Choose what to try.

Explore original experiments, video reviews, and source-based comparisons. Each entry shows its evidence type, date, and commercial disclosure.

Find evidence for your next decision

Newest test, publication, or source-check date first.

4 entries of 35

What the labels mean
Original testsWritten reviews

Tested

DeepSeek V4.1 Flash vs GPT-6 and Claude: Costs, Benchmarks and Failures

We tested DeepSeek V4.1 Flash, GPT-6 Astra and Claude Fable 5.1 on small coding, vision and tool tasks. Here are the exact failures, charges and launch-benchmark limits.

Relationship not stated

Scope and disclosure

The same eight DeepSeek trials appear in both comparison articles; the articles are not independent replications.

A commercial relationship is not stated in this record. This does not establish that the review is independent.

Read the test Download evidence
Original testsWritten reviews

Tested

DeepSeek V4.1 Flash vs Qwen3.8-27B: Coding, Vision and Real Cost

Qwen passed 8/8 trials; DeepSeek passed 7/8 at lower cost. Our small hosted test covers coding, chart reading, tool recovery and the separate local-hardware decision.

Relationship not stated

Scope and disclosure

The same eight DeepSeek trials appear in both comparison articles; the articles are not independent replications.

A commercial relationship is not stated in this record. This does not establish that the review is independent.

Read the test Download evidence
Original testsWritten reviews

Published

Small Qwen3 models on an M4 Pro: a JSON reliability pilot

Three exact Qwen3 files, four short JSON tasks, and saved outputs from an M4 Pro. Inspect strict passes, formatting failures, timings, and sampled process memory.

Relationship not stated

Scope and disclosure

A narrow local pilot: four JSON tasks repeated three times per model. Different quantizations limit model-size comparisons. The 18 retained summaries are unscored.

A commercial relationship is not stated in this record. This does not establish that the review is independent.

Read the test Download evidence
Original testsProject evidence

Tested

Animate a slow push-in on an exact text card

Proof Lab 01: six August 26 clips reassessed September 19. Four strict failures, two unresolved; no certified passes. Inspect the retained outputs, prompt and reproduction files. No new runs.

Relationship not stated

Scope and disclosure

Inspect the saved attempts, method, and limits before applying these results to another task.

A commercial relationship is not stated in this record. This does not establish that the review is independent.

Read the test Download evidence

Read the evidence in context

What the labels mean

Original tests
Kingy-run experiments with saved outputs and a documented method. Small samples and failed attempts stay visible. Some articles share the same runs; entry counts are not independent experiment counts.
Video reviews & demos
Recorded reviews or workflows with an evidence companion. Demonstrated behavior and sourced claims have different limits; these entries are not automatically controlled benchmarks.
Source analysis
Comparisons and buyer reviews based on named sources. They may link to hands-on work, but the source analysis itself is not counted as a new experiment.

Commercial disclosure is a separate filter. “No product sponsorship disclosed” reflects an explicit statement; a video may still include an advertisement. “Relationship not stated” is not a claim of independence. Dates describe the test, publication, or source check shown, not a fresh review of every claim.

This library brings together existing documented records and selected source reviews. The wider archives remain available below.