🏗️ Building on HF

AxionLab

AxionLab-official

111 42 342

AI & ML interests

Owner of SupraLabs and iGPU Lover

Recent Activity

new activity about 9 hours ago

DedeProGames/Kiyo-135M-0960:Kiyo-135M-0960 and Kiyo-135M are byte-identical to SmolLM2-135M intermediate checkpoints (step-960000 and step-240000)

reacted to DavidAU's post with 😔 about 9 hours ago

Qwen 3.5 9B - The Defiant, 27B power ; now with Qwen 3.8 Reasoning modes. 640 ARC-C for both 8bit and 4bit. Model exceeds 7 of 7 benchmarks for Qwen 3.5 9B, Qwen3.5 27B, Qwen3.6 35B-A3B, and meets Qwen 3.6 27B in some cases... and it does so in 4bit and 8bit. Regular and MTP (fast) NEO IMATRIX GGUFs provided. (this model is part of the Qwen 3.6 27B Fable Fusion 711 pipelines: 2200+ likes, 3 million + downloads) NEW - Qwen 3.8 Reasoning Modes: 2 MTP quants (Q6/Q8) Now with 5 reasoning modes (2 new - Spoon / Einstein), and 5 instruct modes (2 new - Spoon / Einstein, all use ZERO REASONING TOKENS) all switchable on the fly via API, direct and "in chat" (yes - model control at the chat/message level). Model name has "plusIQ" in the name. (there is also a extra robust "tools" version too.) https://huggingface.co/DavidAU/Qwen3.5-9B-The-Defiant-Fable-Uncensored-Heretic-NEO-IMATRIX-MAX-MTP-GGUF PS: I have posted a 22k output example using "spoon" mode too.

liked a model about 11 hours ago

TobiasLogic/cagliostro-v3

View all activity

Organizations

New activity in DedeProGames/Kiyo-135M-0960 about 9 hours ago

Kiyo-135M-0960 and Kiyo-135M are byte-identical to SmolLM2-135M intermediate checkpoints (step-960000 and step-240000)

➕👀 10

#5 opened 2 days ago by

TobiasLogic

reacted to DavidAU's post with 😔 about 9 hours ago

Post

2021

Qwen 3.5 9B - The Defiant, 27B power ; now with Qwen 3.8 Reasoning modes.

640 ARC-C for both 8bit and 4bit. Model exceeds 7 of 7 benchmarks for Qwen 3.5 9B, Qwen3.5 27B, Qwen3.6 35B-A3B, and meets Qwen 3.6 27B in some cases... and it does so in 4bit and 8bit. Regular and MTP (fast) NEO IMATRIX GGUFs provided. (this model is part of the Qwen 3.6 27B Fable Fusion 711 pipelines: 2200+ likes, 3 million + downloads)

NEW - Qwen 3.8 Reasoning Modes: 2 MTP quants (Q6/Q8) Now with 5 reasoning modes (2 new - Spoon / Einstein), and 5 instruct modes (2 new - Spoon / Einstein, all use ZERO REASONING TOKENS) all switchable on the fly via API, direct and "in chat" (yes - model control at the chat/message level). Model name has "plusIQ" in the name.

(there is also a extra robust "tools" version too.)

DavidAU/Qwen3.5-9B-The-Defiant-Fable-Uncensored-Heretic-NEO-IMATRIX-MAX-MTP-GGUF

PS: I have posted a 22k output example using "spoon" mode too.

3 replies

liked a model about 11 hours ago

TobiasLogic/cagliostro-v3

Updated 5 minutes ago • 7

reacted to Datdanboi25's post with 🔥❤️ 2 days ago

Post

2833

100 likes on the Open SLM Leaderboard 🎉

176 models, 54 orgs, 5 benchmarks, and a whole community of support!

Thanks to everyone who’s contributed models, reported issues, suggested benchmark improvements, or used the leaderboard to compare and evaluate small language models.

It’s been awesome watching the leaderboard grow into a broader community resource for transparent and reproducible SLM evaluation.

Thank you all, and more to come 🚀

1 reply

New activity in AxiomicLabs/Open_SLM_Leaderboard 4 days ago

Could you please add my following models.

#142 opened 4 days ago by

DedeProGames

reacted to SeaWolf-AI's post with 🚀🔥 4 days ago

Post

5547

Instead of making the fly brain play games, we measured what it is for

Since the Drosophila connectome was released, people have had the fly brain doomscroll a feed, play Beat Saber, drive in GTA. Those demos show that the brain runs. We wanted to show what it is for.

So we gave it a looming object — one of the few things a fly brain is unambiguously built to detect — then deleted a single cell type and repeated the identical stimulus. Remove LC4, 126 cells out of 173,023, and the escape signal falls from 0.840 to 0.091. Eighty-nine percent of the danger signal is gone while the other 172,897 neurons run exactly as before.

Deleting neurons does not do this on its own, which is the whole point of the controls. LC11 is the same class and larger than LC4 — 143 cells and 9,940 outgoing connections against 126 and 7,846 — and removing every one of them changes the signal by 0.000000, to six decimal places. It has to be those 126.

No server and no GPU: a looming stimulus drives fewer than one percent of neurons above threshold, so the whole thing is 40 KB gzipped and runs in your browser.

Try deleting one yourself →
VIDraft/fruitfly-brain

The wiring is the measured connectome, but synaptic strength is a uniform count-based value and the dynamics are a firing-rate model of our choosing — a total-effect measurement of a model, not a recording from a fly. Male CNS connectome, FlyEM / HHMI Janelia with Google Research, Columbia and Harvard (2026), CC BY.

upvoted an article 5 days ago

Article

How Did We Get Here?

NILKNARFGonzo

•

5 days ago

• 7

liked 2 Spaces 6 days ago

CPU LM Speed Leaderboard

⚡

Benchmark base LMs (10M–1B) on CPU — speed and quality

BananaMind SLM Leaderboard

👁

Explore and compare small language model performance

liked a model 7 days ago

GODELEV/Rose-1.5-Medium

98.6M • Updated 8 days ago • 636 • 19

New activity in slmconsortium/README 7 days ago

New Member Ratification: Glint Research

#9 opened 11 days ago by

CompactAI

liked a model 8 days ago

CodeSoft/MetaDiffusion-150M-exp

Text Generation • 0.2B • Updated Aug 7 • 173 • 4

liked a Space 8 days ago

MiniCPM5 WebGPU

🚀

Chat and create websites, docs, sheets & PDFs on your GPU

New activity in AxiomicLabs/Open_SLM_Leaderboard 9 days ago

New License request

👀 2

#131 opened 10 days ago by

KlondikeDev

reacted to Hoglet-33's post with 😔 10 days ago

Post

3176

Today, we planned to release Pebble-50M and Pebble-50M-Chat to the world. Unfortunately, due to a few issues, that didn't go quite as planned.

What happened:
- Some data and benchmark results were lost or corrupted
- The models performed worse on benchmarks than our other Pebble models

Despite that, you can still find both models here:
Pebble-50M-beta: basically-experimental/Pebble-50M-beta
Pebble-50M-Chat-beta: basically-experimental/Pebble-50M-Chat-beta

There are still some interesting improvements in these models:
- Compatible with non-CUDA devices
- Vocabulary increased to 16K tokens
- Context length increased to 16K tokens

For now, there won't be any more Pebble releases for a while. We're going to take some time to experiment with other approaches and hopefully make the next generation a monumental leap over this one.

Follow for updates:
@Hoglet-33