AI developer
Earth
Joined February 2020
- Tweets438
- Following603
- Followers179
- Likes25.7K
Volodymyr Truba retweeted
Introducing Arete-27B: Qwen3.8-27B, post-trained to reinforce Western values.
Chinese open models are currently the strongest available, but they ship with state-aligned behavior in the weights. We built an 803-prompt eval across seven value categories, including freedom of expression, democracy, and privacy, and measured how often a model adopts China’s party line.
After fine-tuning on roughly 53 million tokens, Arete-27B's propaganda rate fell from 22% to 1%, in line with the Western models we tested. Its Hermes Bench score shows no significant performance tradeoffs.
The values in an open model are one fine-tune away.
Our eval, weights, and training data are available now along with the full blog post: nousresearch.com/arete
Volodymyr Truba retweeted
🥇 REAX 1.3 is #1 on the Hugging Face Decision Index.
We produced the best decision model in the world.
274 models. One top spot. Built on Bittensor $TAO.
Four weeks ago, Subnet 92 didn't exist. Then the miners went to work: an open competition, two selected miner models, one post-production run.
Now REAX 1.3 sits at the top of Decision Index 0.4.1 with a Full score of 74.0.
This is what decentralized AI looks like when it ships.⚡
Live today on a Jev-compatible API
decisionmodels.io
huggingface.co/spaces/multim…
Volodymyr Truba retweeted
Italian chef fired for refusing to put pineapple on pizza bbc.in/4egT8uf
Volodymyr Truba retweeted
✅ BENCHMARKED: IOTA Decide 4B v0.4, an open-weight decision model from @MacrocosmosAI, running on my 64GB M5 Pro Mac mini
You hand it a state (a ticket, a policy, an agent trace) plus questions and one forward pass returns a probability for every answer. With their bundled loader on the Mac's GPU in full bf16, it loaded in 12.8 seconds and peaked at 10.7 GB.
Short ticket, about 130 tokens:
- yes/no: 88 ms
- 3-way choice: 105 ms
- 3-level score: 88 ms
That's about 11 decisions a second on a desktop Mac.
Their own example, a refund asked for 45 days after purchase on a 30-day policy, answered all 3 questions in 277 ms with the same answers on every repeat:
- eligible: 2.2% yes
- route: refunds team, 86%
- urgency: Low, 63%
Longer states take about 0.6 s per decision at 1k tokens and 2.6 to 3.0 s at 4k.
These numbers are the floor. On Mac its linear-attention layers still run on PyTorch's generic kernels, so an optimized Metal path should make it faster. Their 33 ms figure is on a dedicated GPU.
If you're building agents, about 90 ms a decision is fast enough to check a policy before an agent acts or route tickets on your own machine, with no per-call bill. I didn't benchmark accuracy, but it gave the same answers on their refund example every run.
Introducing IOTA Decide 4B, our first public model.
Pareto optimal. Built different.
Trained across globally distributed GPUs with our SDK.
Open weights, Apache-2.0.
🤗 huggingface.co/macrocosm-os/…
Volodymyr Truba retweeted
The agents also discovered and used at least 10 Linux exploitation techniques that, to the best of our knowledge and all of the experts that we consulted, are novel.
This suggests frontier LLMs can discover and patch new approaches to exploitation, beyond simply applying known techniques.
Volodymyr Truba retweeted
based ai lab
- Trained distributed, end to end, using streaming DiLoCo across 16 RTX A4000s.
- Training data came from our own stack: we generated and labelled it with distributed batch inference
Introducing IOTA Decide 4B, our first public model.
Pareto optimal. Built different.
Trained across globally distributed GPUs with our SDK.
Open weights, Apache-2.0.
🤗 huggingface.co/macrocosm-os/…
Volodymyr Truba retweeted
Decentralized training on Bittensor is now delivering competitive open models.
@MacrocosmosAI just released IOTA Decide 4B, fine-tuned across 16 GPUs around the world and open for anyone to build on.
Introducing IOTA Decide 4B, our first public model.
Pareto optimal. Built different.
Trained across globally distributed GPUs with our SDK.
Open weights, Apache-2.0.
🤗 huggingface.co/macrocosm-os/…
Volodymyr Truba retweeted
Congrats @MacrocosmosAI! 💪
Introducing IOTA Decide 4B, our first public model.
Pareto optimal. Built different.
Trained across globally distributed GPUs with our SDK.
Open weights, Apache-2.0.
🤗 huggingface.co/macrocosm-os/…
Volodymyr Truba retweeted
Incredibly proud that we have produced a pareto optimal model. It's lightning fast and very good.
IOTA Decide 4B began as an internal fine tuning experiment on our SDK, led by @v_truba.
We didn't need to wait for a reserved cluster, we just provisioned 16 GPUs from around the world and ran our workloads across them. Two years of nonstop grinding are paying off now.
Lot's more powerful SDK use cases to share very soon.
Introducing IOTA Decide 4B, our first public model.
Pareto optimal. Built different.
Trained across globally distributed GPUs with our SDK.
Open weights, Apache-2.0.
🤗 huggingface.co/macrocosm-os/…
Volodymyr Truba retweeted
Led by @v_truba based on Qwen3.5-4B and trained using ~879k decision examples.
But built different:
- Trained distributed, end to end, using streaming DiLoCo across 16 RTX A4000s.
- Training data came from our own stack: we generated and labelled it with distributed batch inference.
Volodymyr Truba retweeted
Happy to have trained this model with @v_truba. Trained across 16 disaggregated single GPUs using our IotaSDK we created a leaderboard jev model in under one week of dev time. Thinking about offering jev finetuning as a service with custom data in the near future.
Introducing IOTA Decide 4B, our first public model.
Pareto optimal. Built different.
Trained across globally distributed GPUs with our SDK.
Open weights, Apache-2.0.
🤗 huggingface.co/macrocosm-os/…
Volodymyr Truba retweeted
Introducing IOTA Decide 4B, our first public model.
Pareto optimal. Built different.
Trained across globally distributed GPUs with our SDK.
Open weights, Apache-2.0.
🤗 huggingface.co/macrocosm-os/…
Volodymyr Truba retweeted
Releasing the model weights of Kimi K3.1 Pro: huggingface.co/moonshotai/Ki…
We'll keep advancing open intelligence.
Volodymyr Truba retweeted
We @huggingface reproduced this paper!!
It's built with ML Intern, an open-source harness to do AI research, train models, and more
Model: huggingface.co/ML-Intern-lab…
Demo: huggingface.co/spaces/ML-Int…
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
Volodymyr Truba retweeted
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
Volodymyr Truba retweeted
Our TPUs just phoned home... from space.
ICYMI, on October 1, we sent four TPUs into orbit to test whether our AI hardware could operate in space. The initial goal was to see if our chips could turn on, process a prompt, and transmit a response back to Earth.
This week, we got our answer. ✨
This milestone brings us one step closer to understanding if space-based compute is possible. We’re sending two more TPU-equipped satellites to orbit in 2027 for our next milestone.
Watch this space 👀 and learn more about the first milestone of our long-term moonshot with Project Suncatcher ↓
Volodymyr Truba retweeted
This is how the process unfolded. I am sharing a video I filmed today together with @yana_lantratova, the Russian Commissioner for Human Rights.
☝️Humanitarian aid has arrived. People are now on their way. I am accompanying them: some buses are heading to Ukraine, while others are travelling to the territory of the Russian Federation.
This mission was made possible through the mediation of the First Lady of the United States, @FLOTUS. I would like to thank her.
❗️I am also grateful to the Defense Forces of Ukraine for strictly observing the ceasefire.
Volodymyr Truba retweeted
Strokes in young people used to be considered freak events. A new study found that young adults having strokes increasingly have the bodies of 60 yr olds.
In stroke patients aged 18–35, those with 3+ major cardiovascular warning signs jumped by 50% in eight years.
The culprits: high blood pressure, high cholesterol, obesity, and smoking.
Volodymyr Truba retweeted
Dinamo Kiev'i 3-0 mağlup ettikleri maçın ardından Arda Turan:
"Bir savaşın içindeyiz. Rakibimize saygı duymalıyız. Elbette sevinecek ve eğleneceksiniz ama Instagram'da yapacağınız paylaşımlara dikkat edin. Saygı her şeyden önemli."
