DaisyChainAI/DaisyChain-Train
Updated โข 5
AI & ML interests
None defined yet.
Recent Activity
View all activity
Quazim0t0ย
updated a
model about 21 hours ago
Post
4947
๐ผ DaisyChain-Web: train a language model with friends or by yourself with multiple devices, in the browser, no install
Open a webpage, share a room link, and every device that joins becomes part of the training cluster. Phones, laptops, old PCs: they connect peer-to-peer over WebRTC and train one shared transformer together, entirely in the browser.
What's actually happening under the hood:
๐ง A mini transformer LM trains on FineWeb-Edu, streamed live from the HuggingFace Hub. Each device pulls its own slice (data parallelism), tokenized with our 16.5k-token Spikewhale tokenizer
โก Every single multiply runs through verified INT8 neural units, no float fallback. On WebGPU browsers it uses the GPU's DP4A integer dot-product hardware, admitted only after proving bit-identical results against the verified units, with a 3รINT8 fast-accurate scheme (CUTLASS's 3xTF32 trick, ported to 8-bit)
๐ Devices average gradients every step under a sync guard: a per-step roster protocol plus weight-hash verification keeps every device's model bit-identical. If anything drifts, training stops instead of silently forking
๐ Live logs show exactly what every device contributes, step by step
๐พ When you're done: test generations right on the page, download a checkpoint, or grab the inference kit, a single self-contained HTML file with the weights baked in that runs generations offline, anywhere
Works solo too. Every extra device just grows the effective batch.
๐ Try it: Quazim0t0/DaisyChain-Web
๐ Training framework: DaisyChainAI/DaisyChain-Train
Updates:
- Block-scaled INT8 quantization
- Batched attention GEMM
- Fused dequant+ReLU epilogue
- Weight-tied unembedding
- WebSocket relay fallback
- Server keepalive ping/pong every 30s
- disconnected-state redial
- Visibility/network-change reconnect
Open a webpage, share a room link, and every device that joins becomes part of the training cluster. Phones, laptops, old PCs: they connect peer-to-peer over WebRTC and train one shared transformer together, entirely in the browser.
What's actually happening under the hood:
๐ง A mini transformer LM trains on FineWeb-Edu, streamed live from the HuggingFace Hub. Each device pulls its own slice (data parallelism), tokenized with our 16.5k-token Spikewhale tokenizer
โก Every single multiply runs through verified INT8 neural units, no float fallback. On WebGPU browsers it uses the GPU's DP4A integer dot-product hardware, admitted only after proving bit-identical results against the verified units, with a 3รINT8 fast-accurate scheme (CUTLASS's 3xTF32 trick, ported to 8-bit)
๐ Devices average gradients every step under a sync guard: a per-step roster protocol plus weight-hash verification keeps every device's model bit-identical. If anything drifts, training stops instead of silently forking
๐ Live logs show exactly what every device contributes, step by step
๐พ When you're done: test generations right on the page, download a checkpoint, or grab the inference kit, a single self-contained HTML file with the weights baked in that runs generations offline, anywhere
Works solo too. Every extra device just grows the effective batch.
๐ Try it: Quazim0t0/DaisyChain-Web
๐ Training framework: DaisyChainAI/DaisyChain-Train
Updates:
- Block-scaled INT8 quantization
- Batched attention GEMM
- Fused dequant+ReLU epilogue
- Weight-tied unembedding
- WebSocket relay fallback
- Server keepalive ping/pong every 30s
- disconnected-state redial
- Visibility/network-change reconnect
- 9 replies
Quazim0t0ย
published a
model 6 days ago
Post
4239
I never really posted about my DaisyChain project because it's still work in progress. I decided to post a small bit about it and the demo.
DaisyChain Genomics: four small DNA/RNA specialists chained behind a learned router that behave like one big genomics model, at ~7ร less active compute.
I built a modular genomics model chasing a 500M-parameter foundation model, then caught myself measuring it wrong. Here's the honest version.
DaisyChain is a different bet: instead of one monolithic DNA model, it's four ~74M specialists (eukaryote, prokaryote, mRNA, splice) chained behind a learned router, each distilled per-domain from HuggingFaceBio's Carbon-500M. Every specialist reports how surprised it is (bits/base) and the router hands each sequence to the link most at home with it. In lineage it's a cluster Branch-Train-Merge mixture of experts, so you can chain on a new domain without retraining the others.
The pitch: ~295M total params (under Carbon-500M), but only one ~74M specialist runs per query, so ~7ร cheaper per token, routing at 100% held-out.
The mistake: Carbon works in 6-mers, and I'd been scoring likelihood as 6-mer cross-entropy. By that number I was +0.043 bits/base behind, splice even "beating" Carbon. But Carbon scores at the base-pair level, which is harder and more honest. Re-run their way:
Real gap: 1.862 vs 1.787 bits/base, +0.089 behind, not +0.043
No domain actually beats Carbon; the "splice win" was an artifact
Seq recovery: euk 31.5% vs 38.9%, bacteria 40.9% vs 54.1%
DaisyChain is still behind Carbon-500M (itself a draft model, not built to top benchmarks), but by a number I can defend, and the gap closes with every per-domain pass. ๐ผ
DaisyChainAI
DaisyChainAI/Daisychain-Genomics-Demo
DaisyChainAI/daisychain-genomics
DaisyChain Genomics: four small DNA/RNA specialists chained behind a learned router that behave like one big genomics model, at ~7ร less active compute.
I built a modular genomics model chasing a 500M-parameter foundation model, then caught myself measuring it wrong. Here's the honest version.
DaisyChain is a different bet: instead of one monolithic DNA model, it's four ~74M specialists (eukaryote, prokaryote, mRNA, splice) chained behind a learned router, each distilled per-domain from HuggingFaceBio's Carbon-500M. Every specialist reports how surprised it is (bits/base) and the router hands each sequence to the link most at home with it. In lineage it's a cluster Branch-Train-Merge mixture of experts, so you can chain on a new domain without retraining the others.
The pitch: ~295M total params (under Carbon-500M), but only one ~74M specialist runs per query, so ~7ร cheaper per token, routing at 100% held-out.
The mistake: Carbon works in 6-mers, and I'd been scoring likelihood as 6-mer cross-entropy. By that number I was +0.043 bits/base behind, splice even "beating" Carbon. But Carbon scores at the base-pair level, which is harder and more honest. Re-run their way:
Real gap: 1.862 vs 1.787 bits/base, +0.089 behind, not +0.043
No domain actually beats Carbon; the "splice win" was an artifact
Seq recovery: euk 31.5% vs 38.9%, bacteria 40.9% vs 54.1%
DaisyChain is still behind Carbon-500M (itself a draft model, not built to top benchmarks), but by a number I can defend, and the gap closes with every per-domain pass. ๐ผ
DaisyChainAI/Daisychain-Genomics-Demo
DaisyChainAI/daisychain-genomics
Post
2889
๐งฉ Verified neural units, now for memory & storage
Following up on my neural-aarch64-units (small MLPs that emulate CPU datapath slices, verified bit-exact over their entire finite input domain โ N/N), I applied the same discipline to memory and storage. Three new repos:
๐ท neural-ddr โ verified units emulating DDR5 logic: DBI (256/256, 512/512), ADDR_MAP (4096/4096), CMD_DECODE (32/32), WR_CRC (512/512), and on-die ECC ODECC (256/256, 3328/3328). Composed into a bridge that presents DDR5 behavior over real DDR3/DDR4 RAM โ flip a bit in every stored byte, ECC corrects all of them.
๐ค Quazim0t0/neural-ddr ยท ๐ป https://github.com/quzi93/neural-ddr
๐๏ธ neural-storage โ a self-healing vault on a neural-verified GF(2โธ) core (LOG/EXP compose to a multiply verified over all 65,536 pairs). Content-addressed dedup + Reed-Solomon so any k of n shards rebuild the whole, plus a whole-drive โ self-healing .pt imager.
๐ค Quazim0t0/neural-storage ยท ๐ป https://github.com/quzi93/neural-storage
๐ฟ neural-cd-preserve โ scan a disc into a self-healing .pt that detects (per-shard SHA-256) and repairs bit-rot, restoring bit-exact even from a damaged copy. Beyond the RS limit it's flagged LOST, never silently wrong.
๐ค Quazim0t0/neural-cd-preserve ยท ๐ป https://github.com/quzi93/neural-cd-preserve
Build your own: golden finite function โ enumerate the domain (decompose big/linear ops like CRC/ECC/GF into bit/byte slices) โ train a small MLP โ verify must be bit-exact on 100% of inputs or it's rejected โ compose. Every repo ships the training + exhaustive-verification scripts.
Honest by construction: dedup removes redundancy, erasure coding adds it, ECC corrects faults โ none of it pretends to beat entropy. Runs on modest/older hardware. ๐ค
Following up on my neural-aarch64-units (small MLPs that emulate CPU datapath slices, verified bit-exact over their entire finite input domain โ N/N), I applied the same discipline to memory and storage. Three new repos:
๐ท neural-ddr โ verified units emulating DDR5 logic: DBI (256/256, 512/512), ADDR_MAP (4096/4096), CMD_DECODE (32/32), WR_CRC (512/512), and on-die ECC ODECC (256/256, 3328/3328). Composed into a bridge that presents DDR5 behavior over real DDR3/DDR4 RAM โ flip a bit in every stored byte, ECC corrects all of them.
๐ค Quazim0t0/neural-ddr ยท ๐ป https://github.com/quzi93/neural-ddr
๐๏ธ neural-storage โ a self-healing vault on a neural-verified GF(2โธ) core (LOG/EXP compose to a multiply verified over all 65,536 pairs). Content-addressed dedup + Reed-Solomon so any k of n shards rebuild the whole, plus a whole-drive โ self-healing .pt imager.
๐ค Quazim0t0/neural-storage ยท ๐ป https://github.com/quzi93/neural-storage
๐ฟ neural-cd-preserve โ scan a disc into a self-healing .pt that detects (per-shard SHA-256) and repairs bit-rot, restoring bit-exact even from a damaged copy. Beyond the RS limit it's flagged LOST, never silently wrong.
๐ค Quazim0t0/neural-cd-preserve ยท ๐ป https://github.com/quzi93/neural-cd-preserve
Build your own: golden finite function โ enumerate the domain (decompose big/linear ops like CRC/ECC/GF into bit/byte slices) โ train a small MLP โ verify must be bit-exact on 100% of inputs or it's rejected โ compose. Every repo ships the training + exhaustive-verification scripts.
Honest by construction: dedup removes redundancy, erasure coding adds it, ECC corrects faults โ none of it pretends to beat entropy. Runs on modest/older hardware. ๐ค
- 1 reply
Post
5398
Disabling Gated Access for some of my models today. I will update this post with the list as I go. I had to go back recently and make updates to a lot of models and bit off more than I could chew with managing all the releases. I didn't realize many of you asked for access and I apologize for not accepting your access to the models you were wanting to look at. I don't like to release something fully unless I feel I put what I could into it for the moment. Some models will remain on gated access, but I will now be accepting those who request to view the repo.
Disabled Gated Access:
Quazim0t0/Byrne-VLM-131M - v2 Updates + Training Instructions
Quazim0t0/Byrne-Speech - 12M Tiny Speech model
Quazim0t0/Byrne-ASR-English - 12M Tiny ASR Model
Quazim0t0/Byrne-VE - Byrne-VE โ Tiny Self-Distilled Vision Encoder (39M)
Quazim0t0/Positronic-144M - Research Artifact
Quazim0t0/SpikeWhale-SNN-216M - Research Artifact
Quazim0t0/Mycel-LM-79M - Research Artifact
Quazim0t0/Chimera-64M - Research Artifact
Accepting Gated Access Requests (7/9):
https://huggingface.co/Quazim0t0/Wheeler-63M
Also uploaded my Neural Photonic Project:
Three trained nets in series: light interferes through the MZI2.pt optical core (verified 256/256), is measured by the PD.pt neural photodetector (verified 1024/1024), and folded into a single OUTPUT byte by the real ADC8 neural-CPU adder. Every value below is computed end-to-end by the three loaded, verified nets โ no analytic formulas.
Demo: https://quazim0t0-neural-photonic-hybrid.hf.space/
Model Weights: Quazim0t0/neural-photonic
AND!
A work in progress:
Ashen Depths
https://quazim0t0-ashendepths.static.hf.space/index.html
Disabled Gated Access:
Quazim0t0/Byrne-VLM-131M - v2 Updates + Training Instructions
Quazim0t0/Byrne-Speech - 12M Tiny Speech model
Quazim0t0/Byrne-ASR-English - 12M Tiny ASR Model
Quazim0t0/Byrne-VE - Byrne-VE โ Tiny Self-Distilled Vision Encoder (39M)
Quazim0t0/Positronic-144M - Research Artifact
Quazim0t0/SpikeWhale-SNN-216M - Research Artifact
Quazim0t0/Mycel-LM-79M - Research Artifact
Quazim0t0/Chimera-64M - Research Artifact
Accepting Gated Access Requests (7/9):
https://huggingface.co/Quazim0t0/Wheeler-63M
Also uploaded my Neural Photonic Project:
Three trained nets in series: light interferes through the MZI2.pt optical core (verified 256/256), is measured by the PD.pt neural photodetector (verified 1024/1024), and folded into a single OUTPUT byte by the real ADC8 neural-CPU adder. Every value below is computed end-to-end by the three loaded, verified nets โ no analytic formulas.
Demo: https://quazim0t0-neural-photonic-hybrid.hf.space/
Model Weights: Quazim0t0/neural-photonic
AND!
A work in progress:
Ashen Depths
https://quazim0t0-ashendepths.static.hf.space/index.html
Post
176
I built/building a small neural Raytracing/Projective-dynamics (PD)/Physics engine from scratch
The idea I wanted to test: in a projective-dynamics solver, could you replace each hand-derived local constraint projection with a learned one, while keeping the analytic parts (rotations, the global solve) exactly as they are? It's one tiny network, shared across every element and across constraint types through material tokens. A new material isn't a new network, just a new token row. Fluids fall out of the same idea, with water treated as one more token.
A few things held up in testing: one tied projector matched five separate per-material solutions, the neural fluid tracked the exact analytic solver closely on a dam-break sim, and a learned warm-start trimmed solver iterations without touching correctness.
Try it here:
(Projective-dynamics)
Dam Break Demo:
https://quazim0t0-neural-physics-engine-demo.hf.space/
FPS Shooter Demo:
https://quazim0t0-neural-combat-evolved.hf.space/
(Raytracing)
Voxel World:
https://quazim0t0-neural-world.static.hf.space/index.html
(Rigid Body)
https://quazim0t0-ashendepths.static.hf.space/
Model Repo:
Quazim0t0/neural-physics-engine
Quazim0t0/neural-raytracing
Quazim0t0/physgait-weights
The idea I wanted to test: in a projective-dynamics solver, could you replace each hand-derived local constraint projection with a learned one, while keeping the analytic parts (rotations, the global solve) exactly as they are? It's one tiny network, shared across every element and across constraint types through material tokens. A new material isn't a new network, just a new token row. Fluids fall out of the same idea, with water treated as one more token.
A few things held up in testing: one tied projector matched five separate per-material solutions, the neural fluid tracked the exact analytic solver closely on a dam-break sim, and a learned warm-start trimmed solver iterations without touching correctness.
Try it here:
(Projective-dynamics)
Dam Break Demo:
https://quazim0t0-neural-physics-engine-demo.hf.space/
FPS Shooter Demo:
https://quazim0t0-neural-combat-evolved.hf.space/
(Raytracing)
Voxel World:
https://quazim0t0-neural-world.static.hf.space/index.html
(Rigid Body)
https://quazim0t0-ashendepths.static.hf.space/
Model Repo:
Quazim0t0/neural-physics-engine
Quazim0t0/neural-raytracing
Quazim0t0/physgait-weights
Post
2397
Created research language model whose channel-mixing block is not an MLP. It is a differentiable Neighbour-Sensing fungal-colony-growth model: each token is expanded into a colony of hyphal tips that grow in a bounded latent region, sense a shared density field, and steer their own growth โ the "MLP" is replaced by a few differentiable steps of colony growth, read back out into the hidden state.
Quazim0t0/Mycel-LM-79M
Also the original SpikeWhale project โ the one that sparked all the other SpikeWhale related projects. Every spiking primitive here is hand-written in plain PyTorch: the leaky integrate-and-fire (LIF) neuron dynamics, the fast-sigmoid surrogate gradient, and the backprop-through-time training loop. No snntorch, no spikingjelly, no norse, no bindsnet โ the network is a genuine from-scratch SNN.
Quazim0t0/SpikeWhale-SNN-216M
Quazim0t0/Mycel-LM-79M
Also the original SpikeWhale project โ the one that sparked all the other SpikeWhale related projects. Every spiking primitive here is hand-written in plain PyTorch: the leaky integrate-and-fire (LIF) neuron dynamics, the fast-sigmoid surrogate gradient, and the backprop-through-time training loop. No snntorch, no spikingjelly, no norse, no bindsnet โ the network is a genuine from-scratch SNN.
Quazim0t0/SpikeWhale-SNN-216M
- 5 replies
Quazim0t0ย
updated a
model 15 days ago
Post
692
Created a causal language model with a non-standard channel-mixing block. It keeps a conventional transformer backbone for token mixing (attention), but replaces the per-layer MLP with a QuazimotoBlock: a bank of coupled phase oscillators (Kuramoto dynamics) arranged in concentric rings, run for a few differentiable Euler steps and read out through [cos ฮธ, sin ฮธ].
Quazim0t0/Positronic-144M
Quazim0t0/Positronic-144M
Quazim0t0ย
updated a
Space 18 days ago
Quazim0t0ย
updated a
Space 21 days ago
Quazim0t0ย
published a
Space 25 days ago
Quazim0t0ย
published a
Space 26 days ago
Quazim0t0ย
published a
model 26 days ago
Post
176
Here's my Hackathon Entries:
build-small-hackathon/ModuleMind
build-small-hackathon/neural-n64
build-small-hackathon/neural-riscv
build-small-hackathon/neural-gb
build-small-hackathon/neural-doom
build-small-hackathon/Flight-Transit-Agent
build-small-hackathon/HuggingWizards
I released a pretrained model base for the end of the Small Models, Big Adventures Hackathon. It is an 86M parameters model, trained on FineWebEdu. Somewhat customized from the standard model. I used my remaining Modal credits from the Hackathon
Model Weights + Demo:
Quazim0t0/Escarda-86M
build-small-hackathon/Escarda-86M-Chat
build-small-hackathon/ModuleMind
build-small-hackathon/neural-n64
build-small-hackathon/neural-riscv
build-small-hackathon/neural-gb
build-small-hackathon/neural-doom
build-small-hackathon/Flight-Transit-Agent
build-small-hackathon/HuggingWizards
I released a pretrained model base for the end of the Small Models, Big Adventures Hackathon. It is an 86M parameters model, trained on FineWebEdu. Somewhat customized from the standard model. I used my remaining Modal credits from the Hackathon
Model Weights + Demo:
Quazim0t0/Escarda-86M
build-small-hackathon/Escarda-86M-Chat
