Human Intelligence

community

Activity Feed Request to join this org

AI & ML interests

Humans answer your questions.

Recent Activity

wop updated a bucket 21 days ago

Human-Intelligence/Database

wop updated a bucket about 1 month ago

Human-Intelligence/storage

wop updated a bucket about 1 month ago

Human-Intelligence/storage

View all activity

wop

posted an update 16 minutes ago

Post

bench-labs/GCTokenizer-v1 , a multilingual tokenizer which does not require a training corpus

bench-labs developed **GCTokenizer-v1**, which is a multi-lingual tokenizer
Available in four sizes: 32K, 65K, 131K and 262K tokens "S, M, L, XL"
It utilizes an encoding scheme which allows it to handle characters in any language around the world

General (multi lingual)
Consensus (from multiple model tokenizers consensus)
Tokenizer

We included an implementation script too,
built like BPE- it can encode arbitrary text, most of the time, efficiently

wop

posted an update 1 day ago

Post

1604

all models now also run on zerogpu!
BenchLabs/Demo-FAST

Speed measurements:
image ~5s (pixelmodel v5 ~0.5s)
3D ~2s
audio ~3s +(upload)

insane.
Yes, we made a new model: bench-labs/AudioModel-v1
BIG thanks to @TobiasLogic

wop

posted an update 3 days ago

Post

2273

🧪 SlopFinder is here!!

We're building a dataset to study what humans actually consider AI slop.

SlopFinder shows you a random piece of AI-generated text and gives you one simple control: **how slop is it?**
No categories. No complicated forms. Just vote and move on.

Every vote helps build the dataset. 🧩

How does it work?
Samples are pulled from existing datasets, shown anonymously, and collected into our annotation pool. After enough votes, they're exported to Hugging Face for everyone to use.

This is an early MVP, so the dataset is small and the system is still evolving.

Vote here:
https://bench-labs.web.app/slopfinder.html
(refresh page if you want to skip)

Dataset:
bench-labs/slop-classification

@benchlabs

17 replies

wop

posted an update 6 days ago

Post

112

🧩 PixelModel v6 is here! 155M parameters
Try it out on our demo (~15 seconds per image) 256x256 🎉
BenchLabs/Demo

Disclaimer: This model does not produce high quality (4k) and does not follow detailed prompts. Does not have negative prompt. 😔

How long did it take to train?
55 hours across two A100 gpu's 🔥

Model repo:
bench-labs/PixelModel-v6
@benchlabs

3 replies

wop

updated a bucket 21 days ago

Human-Intelligence/Database

234 Bytes

wop

posted an update 24 days ago

Post

186

# One Script, Every Benchmark

Every Bench Labs benchmark had its own eval files. Now there's one script, hosted in the leaderboard Space:

curl -sLO https://huggingface.co/spaces/bench-labs/BenchLabs-Leaderboard/resolve/main/script.py
pip install torch transformers
python script.py --model your/model

It runs the full suite with the official scoring — Effortless (exact-match), Easy (hybrid category-aware), Mid (loglikelihood: acc, acc_norm, soft_score_norm) — and reports every category and subcategory, not just one number.

Output includes leaderboard.json: a ready-to-paste models.json entry. Run the script, paste it, open a PR on the [leaderboard]( bench-labs/BenchLabs-Leaderboard). Done.

bench-labs

30 replies

wop

posted an update 25 days ago

Post

3659

# PixelModel v1

Last month we released PixelModel — a neural network whose weights are literally the pixels of a PNG. It was a toy: 202,752 parameters, welded to 32×32 output, trained on six solid-color swatches. It scored FID 566.84 on the Tiny-T2I-Leaderboard, mostly by producing the same yellow noise for every prompt.

Today we're releasing PixelModel v1. It is 8.5× smaller — 23,747 parameters — and it beats v0 on both benchmark metrics while being trained on 20,000 real MS-COCO caption/image pairs instead of six color swatches. The entire model now fits in a 160×149 PNG.

That image is not a visualization of the model. It is the model. All 23,747 weights, one per pixel.

## links

bench-labs
Blog post [read it here ⇗]( bench-labs/blog)
See us on the [Leaderboard ⇗]( FlameF0X/Tiny-T2I-Leaderboard)
Model card [here]( bench-labs/pixelmodel-v1)

## The catch
A 23K-parameter model does not draw sandwiches. With ~1 parameter per training image, the loss-minimizing behavior is to output the average of all plausible images for a caption — caption-conditioned color, light, and layout statistics. Food prompts come out warm and brown; sky prompts come out cool and bright. That is the ceiling for this size class, and we'd rather show it than crop around it.

# cherry on top 🍒
The model generates 600 images (cpu) in 5 (five) seconds.
Thats 5000 images in 24 seconds on cpu.
The model trained on cpu for just 30 minutes.

wop

updated a bucket about 1 month ago

Human-Intelligence/storage

6.83 MB

wop

posted an update 2 months ago

Post

433

🚀 __Monostep v1__ is up →[Monostep-v1 Demo}( wop/Cosmos-T2-Chat)

A tiny (~16.6M) experimental model that predicts 4 tokens per forward pass instead of one. A Transformer trunk pools the prompt into a single vector, then 4 sequential "slot" heads emit a block of tokens left-to-right — a lightweight take on multi-token prediction.

Trained on GSM8K (GPT-2 tokenizer, 10 epochs). It's small and rough — answers are often wrong — but it's a fun little testbed for block decoding. Weights, config, training curves, and a self-contained inference snippet are all in the repo.

Also wired into the Cosmos T2-Accelerate chat demo, where it streams those 4-token blocks live. 🧪

#multitokenprediction #gsm8k #smallmodels