Volodymyr Truba (@v_truba) profile picture

AI developer

Earth
Joined February 2020
Photo by Volodymyr Truba (@v_truba)Photo by Volodymyr Truba (@v_truba)Photo by Volodymyr Truba (@v_truba)Photo by Volodymyr Truba (@v_truba)Photo by Volodymyr Truba (@v_truba)Photo by Volodymyr Truba (@v_truba)Photo by Volodymyr Truba (@v_truba)Photo by Volodymyr Truba (@v_truba)
Volodymyr Truba retweeted
Researchers have used reinforcement learning to teach a 5-tonne John Deere skid-steer loader to perform complex acrobatic manoeuvres autonomously. Trained in simulation and transferred to a real machine, the system can balance on two wheels, navigate rough terrain, spin, and control its bucket before landing safely. The research demonstrates the potential of AI-driven automation for heavy construction and industrial machinery. Video Credit: @leggedrobotics / @ETH Paper: wevlv.co/4kltYys #engineering #technology
392
1,340
866
13,641
2,115,963
Volodymyr Truba retweeted
Introducing Qwen3.8-27B-pi-adapted 🤗 We adapted Qwen's model specifically for the pi coding agent using reinforcement learning and a LoRA adapter. The results, measured as pass@1: 🚀 SWE-bench Pro (Hard 51): 29.4% → 62.7% 💻 Terminal-Bench 2.1: 39.5% → 55.8% ⚡ 55–62% fewer tool calls per task This is the power of open models: you can adapt them to YOUR tools and YOUR use cases. Let’s build! 💛 Open weights: huggingface.co/harness-hub/Q… Methodology: huggingface.co/blog/harness-…
303
642
172
5,987
448,103
Volodymyr Truba retweeted
GPT-6.2 Astra from @OpenAI on ARC-AGI (Verified): - ARC-AGI-3: 100.0% (SOTA), $4.1K (provider adapter), 70.3%, $24.2K (standard) - ARC-AGI-2: 95.8% (SOTA), $0.50/task - ARC-AGI-1: 98.5%, $0.17/task GPT-6.2 Astra is the first model to achieve a 100% score on ARC-AGI-3.
83
217
122
2,408
331,408
Volodymyr Truba retweeted
The Graceful Tree Conjecture is 59 years old. Our miners proved one of its five-leaf cases. Every tree with one degree-4 vertex, one degree-3 vertex and all other vertices of degree at most 2 has a graceful labeling. Verified in Lean. Paper and proof below.
1
13
1
35
1,059
Volodymyr Truba retweeted
Mercedes stoppt alle weiteren Investitionen in Deutschland – Werksschließungen werden geprüft to.welt.de/tVl9Zoj
976
2,130
862
7,963
1,175,567
Volodymyr Truba retweeted
JUST IN - Jensen Huang resigns as Nvidia CEO.
226
743
354
5,317
395,433
Volodymyr Truba retweeted
🚨 THIS IS NOT A VIDEO MODEL 🚨 This is fable 5.2 with blender and unreal. I told fable to create me some game assets but to really focus on the environment. I went to sleep and woke up to this: I am speechless.
106
88
8
1,331
79,438
Volodymyr Truba retweeted
Claude reverse engineered my 32-year-old Jeep. Then it started controlling it. Edited with Fable 5.2
630
1,778
636
19,393
1,800,610
Volodymyr Truba retweeted
Introducing Gemini Decision 1, our new frontier decision model in the Gemini API. It is $0.10 per 1M input tokens with no additional cost. With support for image, audio, video, and PDF natively, 256K context window, & 1K questions. You can try it for free in the API!
600
636
360
10,636
820,170
Volodymyr Truba retweeted
RedTeam - Bittensor $TAO Subnet 61 theredteam.io At the House of Lords last week:
20
59
28
342
34,737
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
1,853
1,495
1,303
15,087
2,676,975
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…
28
82
45
316
104,484
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/…
1
8
3
57
4,866
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.
10
31
13
728
143,776
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/…
3
8
50
2,104
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/…
4
36
204
9,172
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/…
2
1
36
1,022
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/…
2
9
1
63
2,615
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.
1
1
1
16
1,946