ByteDance's veRL: Volcano Engine Reinforcement Learning for LLM
veRL is a flexible, efficient and production-ready RL training framework designed for large language models (LLMs).
github.com/volcengine/v...
veRL is a flexible, efficient and production-ready RL training framework designed for large language models (LLMs).
github.com/volcengine/v...
GitHub - volcengine/verl: veRL: Volcano Engine Reinforcement Learning for LLM
veRL: Volcano Engine Reinforcement Learning for LLM - volcengine/verl
github.com
January 30, 2025 at 5:10 AM
ByteDance's veRL: Volcano Engine Reinforcement Learning for LLM
veRL is a flexible, efficient and production-ready RL training framework designed for large language models (LLMs).
github.com/volcengine/v...
veRL is a flexible, efficient and production-ready RL training framework designed for large language models (LLMs).
github.com/volcengine/v...
ByteDance's OpenViking: The Context Database for AI Agents
OpenViking unifies the management of context (memory, resources, and skills) that Agents need through a file system paradigm, enabling hierarchical context delivery and self-evolving.
github.com/volcengine/O...
OpenViking unifies the management of context (memory, resources, and skills) that Agents need through a file system paradigm, enabling hierarchical context delivery and self-evolving.
github.com/volcengine/O...
GitHub - volcengine/OpenViking: OpenViking is an open-source context database designed specifically for AI Agents. OpenViking unifies the management of context (memory, resources, and skills) that Age...
OpenViking is an open-source context database designed specifically for AI Agents. OpenViking unifies the management of context (memory, resources, and skills) that Agents need through a file syste...
github.com
February 20, 2026 at 6:13 AM
ByteDance's OpenViking: The Context Database for AI Agents
OpenViking unifies the management of context (memory, resources, and skills) that Agents need through a file system paradigm, enabling hierarchical context delivery and self-evolving.
github.com/volcengine/O...
OpenViking unifies the management of context (memory, resources, and skills) that Agents need through a file system paradigm, enabling hierarchical context delivery and self-evolving.
github.com/volcengine/O...
🔥 Hot Repo！ 🔥 (100+ new stars)
📦 volcengine / verl
⭐ 1,765 (+124)
🗒 Python
veRL: Volcano Engine Reinforcement Learning for LLM
📦 volcengine / verl
⭐ 1,765 (+124)
🗒 Python
veRL: Volcano Engine Reinforcement Learning for LLM
GitHub - volcengine/verl: veRL: Volcano Engine Reinforcement Learning for LLM
veRL: Volcano Engine Reinforcement Learning for LLM - volcengine/verl
github.com
February 2, 2025 at 6:01 PM
🔥 Hot Repo！ 🔥 (100+ new stars)
📦 volcengine / verl
⭐ 1,765 (+124)
🗒 Python
veRL: Volcano Engine Reinforcement Learning for LLM
📦 volcengine / verl
⭐ 1,765 (+124)
🗒 Python
veRL: Volcano Engine Reinforcement Learning for LLM
Everything is fully open-sourced (algorithm, code, dataset, verifier, and model) and built with verl project (github.com/volcengine/v...).
Homepage: dapo-sia.github.io
Paper: dapo-sia.github.io/static/pdf/d...
Code: github.com/volcengine/v...
Homepage: dapo-sia.github.io
Paper: dapo-sia.github.io/static/pdf/d...
Code: github.com/volcengine/v...
DAPO
DAPO: an Open-Source LLM Reinforcement Learning
System at Scale
dapo-sia.github.io
March 17, 2025 at 11:52 PM
Everything is fully open-sourced (algorithm, code, dataset, verifier, and model) and built with verl project (github.com/volcengine/v...).
Homepage: dapo-sia.github.io
Paper: dapo-sia.github.io/static/pdf/d...
Code: github.com/volcengine/v...
Homepage: dapo-sia.github.io
Paper: dapo-sia.github.io/static/pdf/d...
Code: github.com/volcengine/v...
- Improves Retrieval Effect
- Observable Context
- Context Self-Iteration
Repo: github.com/volcengine/O...
- Observable Context
- Context Self-Iteration
Repo: github.com/volcengine/O...
GitHub - volcengine/OpenViking: OpenViking is an open-source context database designed specifically for AI Agents(such as openclaw). OpenViking unifies the management of context (memory, resources, an...
OpenViking is an open-source context database designed specifically for AI Agents(such as openclaw). OpenViking unifies the management of context (memory, resources, and skills) that Agents need th...
github.com
March 17, 2026 at 8:00 PM
- Improves Retrieval Effect
- Observable Context
- Context Self-Iteration
Repo: github.com/volcengine/O...
- Observable Context
- Context Self-Iteration
Repo: github.com/volcengine/O...
May 25, 2026 at 3:13 PM
ByteDance CEO Liang Rubo detailed an AI office stack at the 2026 Feishu Future Unlimited Conference, integrating Doubao, Feishu, and Volcengine into a single operating picture.
Source: Pandaily
Source: Pandaily
ByteDance Ties Doubao, Feishu and Volcengine Into an AI Office Stack
At Feishu's 2026 conference, CEO Liang Rubo cast Doubao agents, Feishu collaboration and Volcengine models-and-compute as one enterprise stack, highlighting Feishu 8.0 Agent CLI endpoints rising from 247 to 767 and H1 ARR growth 2.5 times year earlier.
pandaily.com
September 16, 2026 at 8:40 AM
ByteDance CEO Liang Rubo detailed an AI office stack at the 2026 Feishu Future Unlimited Conference, integrating Doubao, Feishu, and Volcengine into a single operating picture.
Source: Pandaily
Source: Pandaily
March 26, 2026 at 1:23 PM
Seres to launch new auto brand in June with ByteDance involvement, report says
The new brand is targeting the younger and sports-oriented markets and will deeply integrate AI technology from ByteDance's Volcengine. cnev.co/2RZOj9L 👇
The new brand is targeting the younger and sports-oriented markets and will deeply integrate AI technology from ByteDance's Volcengine. cnev.co/2RZOj9L 👇
Seres to launch new auto brand in June with ByteDance involvement, report says
The new brand is targeting the younger and sports-oriented markets and will deeply integrate AI technology from ByteDance's Volcengine.
cnev.co
May 30, 2026 at 8:14 AM
Seres to launch new auto brand in June with ByteDance involvement, report says
The new brand is targeting the younger and sports-oriented markets and will deeply integrate AI technology from ByteDance's Volcengine. cnev.co/2RZOj9L 👇
The new brand is targeting the younger and sports-oriented markets and will deeply integrate AI technology from ByteDance's Volcengine. cnev.co/2RZOj9L 👇
volcengine / verl
verl: Volcano Engine Reinforcement Learning for LLMs
https://github.com/volcengine/verl
verl: Volcano Engine Reinforcement Learning for LLMs
https://github.com/volcengine/verl
March 3, 2025 at 9:15 PM
volcengine / verl
verl: Volcano Engine Reinforcement Learning for LLMs
https://github.com/volcengine/verl
verl: Volcano Engine Reinforcement Learning for LLMs
https://github.com/volcengine/verl
November 27, 2025 at 1:08 PM
At the Tencent Cloud Databases + AI Summit in Shanghai on May 29, Zhu Binbin (Valkey TSC Member) met with Jean Li from Volcengine. They discussed the future roadmap for Valkey and community governance. Even more excitingly, Volcengine plans to launch a Valkey service by the end of Q1 next year!
June 16, 2026 at 4:00 AM
At the Tencent Cloud Databases + AI Summit in Shanghai on May 29, Zhu Binbin (Valkey TSC Member) met with Jean Li from Volcengine. They discussed the future roadmap for Valkey and community governance. Even more excitingly, Volcengine plans to launch a Valkey service by the end of Q1 next year!
September 17, 2026 at 12:00 PM
OSS reinforcement learning lib by ByteDance is used to reproduce DeepSeek R1 Article URL: https:/...
https://github.com/volcengine/verl
Event Attributes
https://github.com/volcengine/verl
Event Attributes
GitHub - volcengine/verl: veRL: Volcano Engine Reinforcement Learning for LLM
veRL: Volcano Engine Reinforcement Learning for LLM - volcengine/verl
github.com
February 6, 2025 at 7:56 PM
OSS reinforcement learning lib by ByteDance is used to reproduce DeepSeek R1 Article URL: https:/...
https://github.com/volcengine/verl
Event Attributes
https://github.com/volcengine/verl
Event Attributes
🚀 Skyrocketing！ 🚀 (200+ new stars)
📦 volcengine / OpenViking
⭐ 29,036 (+239)
🗒 Python
Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
📦 volcengine / OpenViking
⭐ 29,036 (+239)
🗒 Python
Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
GitHub - volcengine/OpenViking: Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills. - volcengine/OpenViking
github.com
August 18, 2026 at 4:17 PM
🚀 Skyrocketing！ 🚀 (200+ new stars)
📦 volcengine / OpenViking
⭐ 29,036 (+239)
🗒 Python
Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
📦 volcengine / OpenViking
⭐ 29,036 (+239)
🗒 Python
Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
AI developer tools for agents, codebases and model access
OpenViking (Python, 37,801 stars): Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
https://github.com/volcengine/OpenViking
OpenViking (Python, 37,801 stars): Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
https://github.com/volcengine/OpenViking
September 19, 2026 at 6:26 PM
AI developer tools for agents, codebases and model access
OpenViking (Python, 37,801 stars): Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
https://github.com/volcengine/OpenViking
OpenViking (Python, 37,801 stars): Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
https://github.com/volcengine/OpenViking
OpenViking treats agent context more like a filesystem than a magic memory layer: viking://, tiered reads, inspectable retrieval paths, CLI and Docker deployment, plus integrations across Codex, Claude Code, Cursor, OpenCode, and MCP.
github.com/volcengine/O...
github.com/volcengine/O...
GitHub - volcengine/OpenViking: Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills. - volcengine/OpenViking
github.com
August 21, 2026 at 12:01 PM
OpenViking treats agent context more like a filesystem than a magic memory layer: viking://, tiered reads, inspectable retrieval paths, CLI and Docker deployment, plus integrations across Codex, Claude Code, Cursor, OpenCode, and MCP.
github.com/volcengine/O...
github.com/volcengine/O...
And more, including custom and local APIs:
March 2, 2025 at 6:57 PM
And more, including custom and local APIs:
用途型: AIエージェント専用の文脈データベースらしい
感想系: メモリやスキルをファイル管理できる発想が便利そう
yug1224 starred volcengine/OpenViking
https://github.com/volcengine/OpenViking
感想系: メモリやスキルをファイル管理できる発想が便利そう
yug1224 starred volcengine/OpenViking
https://github.com/volcengine/OpenViking
GitHub - volcengine/OpenViking: OpenViking is an open-source context database designed specifically for AI Agents(such as openclaw). OpenViking unifies the management of context (memory, resources, and skills) that Agents need through a file system paradigm, enabling hierarchical context delivery and self-evolving.
github.com
March 15, 2026 at 9:58 AM
用途型: AIエージェント専用の文脈データベースらしい
感想系: メモリやスキルをファイル管理できる発想が便利そう
yug1224 starred volcengine/OpenViking
https://github.com/volcengine/OpenViking
感想系: メモリやスキルをファイル管理できる発想が便利そう
yug1224 starred volcengine/OpenViking
https://github.com/volcengine/OpenViking
ByteDance's Volcengine upgrades open-source 'verl' with agentic reinforcement learning and async training. This architecture could redefine large model scaling, enabling efficient multi-step reasoning without expensive GPU clusters. The implications for democratizing advanced agent...
ByteDance-Backed Volcengine Updates 'verl' Framework with Agentic RL and Asynchronous Training
Volcengine has released version 0.5.0 of verl, its distributed reinforcement learning framework, introducing agentic tool-use capabilities and significant throughput optimisations for Chinese AI developers.
aichina.news
March 30, 2026 at 2:39 AM
ByteDance's Volcengine upgrades open-source 'verl' with agentic reinforcement learning and async training. This architecture could redefine large model scaling, enabling efficient multi-step reasoning without expensive GPU clusters. The implications for democratizing advanced agent...
ByteDance's Volcengine integrates Qwen3VL into its reinforcement learning framework with native Huawei Ascend support. This deepens the alignment between advanced multimodal models and domestic silicon, potentially accelerating open-source RL adoption across China's AI ecosystem.
ByteDance’s Volcengine Enhances 'verl' Framework with Qwen3VL Support and Huawei Ascend Compatibility
Volcengine has released version 0.6.1 of its open-source reinforcement learning framework, introducing asynchronous training capabilities and expanded support for Chinese hardware and vision models.
aichina.news
March 30, 2026 at 2:37 AM
ByteDance's Volcengine integrates Qwen3VL into its reinforcement learning framework with native Huawei Ascend support. This deepens the alignment between advanced multimodal models and domestic silicon, potentially accelerating open-source RL adoption across China's AI ecosystem.
今日のGitHubトレンド
volcengine/verl
このリポジトリは、大規模言語モデル（LLMs）のための柔軟で効率的、かつ実運用可能な強化学習（RL）トレーニングライブラリ「verl」を提供する。
これは「HybridFlow」フレームワークのオープンソース実装であり、推論能力やコーディング能力の向上など、多様なタスクにおけるRLHF（人間フィードバックによる強化学習）の適用を目的としている。
volcengine/verl
このリポジトリは、大規模言語モデル（LLMs）のための柔軟で効率的、かつ実運用可能な強化学習（RL）トレーニングライブラリ「verl」を提供する。
これは「HybridFlow」フレームワークのオープンソース実装であり、推論能力やコーディング能力の向上など、多様なタスクにおけるRLHF（人間フィードバックによる強化学習）の適用を目的としている。
GitHub - volcengine/verl: verl: Volcano Engine Reinforcement Learning for LLMs
verl: Volcano Engine Reinforcement Learning for LLMs - volcengine/verl
github.com
November 30, 2025 at 11:15 AM
今日のGitHubトレンド
volcengine/verl
このリポジトリは、大規模言語モデル（LLMs）のための柔軟で効率的、かつ実運用可能な強化学習（RL）トレーニングライブラリ「verl」を提供する。
これは「HybridFlow」フレームワークのオープンソース実装であり、推論能力やコーディング能力の向上など、多様なタスクにおけるRLHF（人間フィードバックによる強化学習）の適用を目的としている。
volcengine/verl
このリポジトリは、大規模言語モデル（LLMs）のための柔軟で効率的、かつ実運用可能な強化学習（RL）トレーニングライブラリ「verl」を提供する。
これは「HybridFlow」フレームワークのオープンソース実装であり、推論能力やコーディング能力の向上など、多様なタスクにおけるRLHF（人間フィードバックによる強化学習）の適用を目的としている。