Haibin
@eric_haibin_lin
Followers
2K
Following
357
Media
27
Statuses
336
Bytedance Seed LLM Systems, @verl_project, author of Megascale. Prev. Amazon AI, ex-committer for BytePS, Gluon-NLP, @ApacheMXNet
Joined November 2012
RT @verl_project: The 1st verl meetup will be held at ICML Vancouver on July 16th! Please join us if you will be there! .
0
2
0
RT @verl_project: If you're in Singapore on 7/11, do not miss this meetup! Talks from the verl community: .- LLMs to optimize code performa….
0
4
0
RT @Agentica_: 🚀 Introducing DeepSWE 🤖: our fully open-sourced, SOTA software engineering agent trained purely with RL on top of Qwen3-32B.….
0
65
0
🚀🚀🚀.
DeepSeek 671b and Qwen3 236b support with Megatron backend is now available as preview in verl v0.4.0 🔥🔥🔥.We will continue optimizing MoE model performance down the road. DeepSeek 671b: .verl v0.4:
0
0
21
RT @InfiniAILab: 🥳 Happy to share our new work – Kinetics: Rethinking Test-Time Scaling Laws. 🤔How to effectively build a powerful reasoni….
0
66
0
RT @DongfuJiang: Introducing VerlTool - a unified and easy-to-extend tool agent training framework based on verl. Recently, there's been a….
0
62
0
RT @verl_project: Multi-GPU LoRA RL is now available in verl! It enables 70B+ model RL with 8 GPUs in bf16. Getting started: https://t.co….
0
9
0
RT @langfengq: "verl-agent" is also the official repo for paper "Group-in-Group Policy Optimization for LLM Agent Training"..
0
1
0
RT @verl_project: As a community we continue to provide popular open source post-training recipes. Recently we worked together with the com….
0
4
0
SkyRL is a great work extending @verl_project with environments for agent tasks. It leverages the sglang multi-turn/tool calling feature recently added to verl:
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
1
28
179
Very efficient yet powerful 8B coding model trained with @verl_project. Check it out! .
🚀 Thrilled to introduce Seed-Coder, our new open-source family of 🔥powerful, 🔍transparent, ⚡parameter-efficient code models at 8B scale!. 💥 Small model, big results!.🥇 BigCodeBench, FullStack Bench, MHPP: impressive results among lightweight open-source models, including
1
10
74
RT @verl_project: verl is embracing @PyTorch fsdp2! Better throughput, memory usage, and composability with torch.compile!. Please try it o….
0
6
0
RT @verl_project: qwen3 with both Megatron and FSDP reinforcement learning support is now available in verl!. https….
0
11
0
RT @verl_project: verl provides day 1 RL support for qwen3 models! Both sequence parallelism and remove padding acceleration are available.….
0
13
0
RT @tongyx361: Welcome to enjoy ICLR Expo Talk Panel "verl: Flexible and Efficient Infrastructures for Post-training LLMs" from ByteDance S….
0
11
0
RT @verl_project: Deploy verl on AMD GPUs for fast, scalable RLHF training with ROCm optimization, docker scripts, and impressive throughpu….
0
4
0
RT @qiying_yu: #ICLR2025.I am going to present VAPO & DAPO twice at ICLR, two SOTA LLM RL algorithms. 1. The 1-2 pm verl Expo Talk, Apr 2….
0
3
0
RT @verl_project: We will present latest updates of verl at #ICLR2025:.- recent RL recipes (DAPO, etc).- RL with tool calling & multi-turn….
0
15
0
Nice work training tool-augmented RL built on top of @verl_project and the DAPO algorithm. And The infrastructure for tool use / multi-turn RL is under review. Feedback is welcome! .
1
25
172
RT @lmsysorg: Hello, everyone! Get ready for an electrifying session this Saturday at 6 PM Pacific Time (9 PM Eastern Time) or Sunday at 9….
0
23
0