TEST RECORDS · open-source AI models, run before written upISSUED MON & THU

EVERY MON & THU

Open-source AI models, tested to spec before you commit.

Setup notes, honest caveats and license checks for image, audio, vision and 3D models. Everything gets run before it gets written up. Two records every Monday and Thursday, free.

113 field reports on record · free

Models

Latest field reports

LayaFIELD NOTE

ISSUE #113 · yesFIELD NOTE

SEP 24, 2026

Answers yes/no and multiple-choice questions about any text, with a confidence score on every answer

yes/no

Reads Text Tables And Formulas Out Of Scanned Pages And Phone Photos Even When The Page Is CurvedFIELD NOTE

FIELD REPORTFIELD NOTE

SEP 24, 2026

Reads text, tables and formulas out of scanned pages and phone photos, even when the page is curved

Confucius4-R2T2AUDIO · VOICE

ISSUE #112AUDIO

SEP 21, 2026

Transcribes speech live, committing each word as soon as it is heard

Qwen-Image-2.1IMAGE · EDIT

ISSUE #111IMAGE

SEP 21, 2026

Generates images with transparent backgrounds and edits photos from up to ten reference shots

K2-Horizon-7BFIELD NOTE

ISSUE #110FIELD NOTE

SEP 17, 2026

Reads an entire book in one prompt and answers questions about it

OUI-1IMAGE · GENERATION

ISSUE #109IMAGE

SEP 17, 2026

Generates working user interfaces the way image models generate pictures

Krea2-Turbo-Distill-4step-LoRAIMAGE · GENERATION

ISSUE #108MODEL

SEP 14, 2026

Make Krea 2 images twice as fast without losing fine detail

AuK-FlashIMAGE · EDIT

ISSUE #107IMAGE

SEP 14, 2026

Edit what a recording says, then change its mood, speed or accent

MiniCPM5-2BFIELD NOTE

ISSUE #105FIELD NOTE

SEP 10, 2026

Solves code and math problems and calls tools while keeping pace with models twice its size

Qwen3.8-Flash-NextVIDEO · MOTION

ISSUE #106VIDEO

SEP 10, 2026

Answers questions about images, hour-long videos and long documents, with a dial for how hard it thinks

Viggle-AnimateIMAGE · EDIT

ISSUE #103IMAGE

SEP 7, 2026

Replace the character in a video by editing just one frame

VibeVoice-ASR-Streaming-7BAUDIO · VOICE

ISSUE #104AUDIO

SEP 7, 2026

Transcribe speech as it happens and label each speaker

12 of 114 records shownFULL ARCHIVE →