NovaSky (@NovaSkyAI) / X

NovaSky

148 posts

NovaSky

@NovaSkyAI

Building SkyRL at @BerkeleySky Join the Slack community: join.slack.com/t/skyrl/shared…

Berkeley, California

github.com/NovaSky-AI/Sky…

Joined January 2025

Following

2,833

Followers

Pinned
NovaSky
@NovaSkyAI
Feb 13
We are excited to announce that SkyRL now implements the Tinker API. Run Tinker training scripts on your own hardware with zero code changes. Try it out today:
Tyler Griggs
@tyler_griggs_
Feb 13
SkyRL now implements the Tinker API. Now, training scripts written for Tinker can run on your own GPUs with zero code changes using SkyRL's FSDP2, Megatron, and vLLM backends. Blog: novasky-ai.notion.site/skyrl-tinker 🧵
novasky-ai.notion.site
SkyRL tx v0.0.3 Release
Philipp Moritz, Tyler Griggs, and the SkyRL Team
2.7K
NovaSky
@NovaSkyAI
Jan 10, 2025
1/6 🚀 Introducing Sky-T1-32B-Preview, our fully open-source reasoning model that matches o1-preview on popular reasoning and coding benchmarks — trained under $450! 📊Blog: novasky-ai.github.io/posts/sky-t1/ 🏋️‍♀️Model weights: huggingface.co/NovaSky-AI/Sky…
312K
NovaSky
@NovaSkyAI
Feb 14, 2025
Interleaving SFT + RL unleashes the power of small models. 🚀 Introducing Sky-T1-7B, a SOTA open-recipe model trained with 4-step SFT->RL->SFT->RL. We also release Sky-T1-mini, trained with RL from DeepSeek-R1-Distill-Qwen-7B model, approaching o1-mini performance on math tasks.
31K
NovaSky
@NovaSkyAI
May 7, 2025
1/N Introducing SkyRL-v0, our RL training pipeline enabling efficient RL training for long-horizon, real-environment tasks like SWE-Bench. We also open-source a series of our early trained models to showcase the potential of end-to-end online RL training on long-horizon (20-50
96K
NovaSky
@NovaSkyAI
Feb 21, 2025
1/8 🚀 Introducing S*: Test-Time Scaling for Code Generation, start of our releases in the coding domain @NovaSkyAI . S* enables (1) non-reasoning models surpass reasoning models: GPT-4o-mini + S* > o1-preview. (2) open models compete SOTA: R1-Distilled-32B +S* ~= o1 (high).
25K
NovaSky
@NovaSkyAI
Jun 26, 2025
✨Release: We upgraded SkyRL into a highly-modular, performant RL framework for training LLMs. We prioritized modularity—easily prototype new algorithms, environments, and training logic with minimal overhead. 🧵👇 Blog: novasky-ai.notion.site/skyrl-v01 Code: github.com/NovaSky-AI/Sky…
44K
NovaSky
@NovaSkyAI
May 22, 2025
1/N Introducing SkyRL-SQL, a simple, data-efficient RL pipeline for Text-to-SQL that trains LLMs to interactively probe, refine, and verify SQL queries with a real database. 🚀 Early Result: trained on just ~600 samples, SkyRL-SQL-7B outperforms GPT-4o, o4-mini, and SFT model
32K
NovaSky
@NovaSkyAI
Jul 10, 2025
🔎 SkyRL + Search-R1 Training a multi-turn search agent doesn’t have to be complicated. With SkyRL, reproducing the SearchR1 recipe at high training throughput is quick and easy! We wrote up a detailed guide to show you how: novasky-ai.notion.site/skyrl-searchr1 1/N 🧵
18K
NovaSky
@NovaSkyAI
Jan 10, 2025
Replying to @NovaSkyAI
4/6 ⚙️ The training recipe: - Base: Qwen2.5-32B-Instruct - Data: Curated from QwQ-32B, enhanced with GPT-4o-mini, reject sampling for high-quality math & coding reasoning traces. - Cost: 8 H100 GPUs, 19 hours, $450.
14K
NovaSky
@NovaSkyAI
Jan 10, 2025
Replying to @NovaSkyAI
2/6📂 Data curation, train, eval code, 17K training data: github.com/NovaSky-AI/Sky… Collaborate, replicate, and innovate! 💡
GitHub - NovaSky-AI/SkyThought: Sky-T1: Train your own O1 preview model within $450
From github.com
13K
NovaSky
@NovaSkyAI
Jan 10, 2025
Replying to @NovaSkyAI
6/6 Acknowledgements: Built with support from: @LambdaAPI @anyscalecompute for compute Academic Insights from STILL-2 & Qwen Teams 💻 Built at Berkeley’s Sky Computing Lab @BerkeleySky with the amazing NovaSky team: Contact: [email protected]!
9K
NovaSky
@NovaSkyAI
Jan 10, 2025
Replying to @NovaSkyAI
5/6🌟 Sky-T1-32B-Preview is just the beginning! Next steps: - Efficient models with strong reasoning - Explore advanced techniques for test-time scaling
9.6K
NovaSky
@NovaSkyAI
Jan 23, 2025
1/5 ⚡️Presenting Sky-T1-32B-Flash⚡️, our open reasoning model that tackles "overthinking" to cut generation lengths (and inference cost!) by 50% without sacrificing accuracy – tuned with only $275! 📊Blog: novasky-ai.github.io/posts/reduce-o… 🏋️‍♀️Weights: huggingface.co/NovaSky-AI/Sky…
13K
NovaSky
@NovaSkyAI
Jan 10, 2025
Replying to @NovaSkyAI
3/6📈 Sky-T1-32B-Preview excels in both math & coding: - Math500: 82.4% (o1-preview: 81.4%) - AIME24: 43.3% (o1-preview: 40.0%) - LiveCodeBench-Hard: 17.9% (o1-preview: 16.3%)
9.9K