ByteDance on SoloSoft

DeerFlow: ByteDance's Open-Source LLM Workflow Engine

Mon, 01 Jan 0001 00:00:00 +0000

Building production LLM applications involves far more than making a single API call. Real-world applications chain multiple LLM calls together, combine them with data processing steps, apply conditional logic, handle errors gracefully, and manage state across the pipeline. DeerFlow by ByteDance provides a comprehensive workflow engine for building these complex LLM applications, with a visual pipeline designer that makes the development process accessible and transparent.

DeerFlow is built on the observation that most LLM applications follow identifiable patterns: retrieve-then-generate (RAG), multi-step reasoning, LLM-as-judge evaluation, and agent-based tool use. Rather than implementing these patterns from scratch each time, DeerFlow provides reusable pipeline components that can be wired together both visually and programmatically.

Seed1.5-VL: ByteDance's Vision-Language Foundation Model Achieving 38 SOTA Benchmarks

Mon, 01 Jan 0001 00:00:00 +0000

In the rapidly advancing field of vision-language models, a new heavyweight has emerged from an unexpected corner. Seed1.5-VL, developed by ByteDance’s Seed team, has achieved state-of-the-art results on an astonishing 38 out of 60 public benchmarks, spanning image understanding, video comprehension, document parsing, and multi-image reasoning.

Built on a 20-billion parameter Mixture-of-Experts (MoE) architecture with approximately 2 billion activated parameters per token, Seed1.5-VL represents a careful balancing act between raw capability and computational efficiency. It outperforms models with far larger parameter counts while maintaining inference speeds suitable for real-world applications.

The model’s benchmark sweep is remarkable not just for the number of wins, but for the breadth of categories it dominates. From OCR and chart understanding to multi-image reasoning and video comprehension, Seed1.5-VL demonstrates that ByteDance’s research team has achieved something genuinely comprehensive in the multimodal space.

StoryDiffusion: Consistent Self-Attention for Long-Range Image and Video Generation

Mon, 01 Jan 0001 00:00:00 +0000

StoryDiffusion is a research project from Nankai University and ByteDance that tackles one of the hardest problems in generative AI: maintaining visual consistency across long sequences of images and videos. Accepted as a major research contribution, it introduces a novel consistent self-attention (CSA) mechanism that enables diffusion models to generate coherent comic strips, animations, and videos – all without finetuning or per-sequence training.

The core challenge StoryDiffusion addresses is simple to state but extremely difficult to solve: how do you generate a sequence of images where the same character looks consistently the same in every frame? Previous diffusion models could produce stunning single images, but when asked to generate a multi-panel comic or a video clip, characters would subtly change appearance between frames – a different nose shape, a changed outfit, a shifted background style.

VeRL: ByteDance's Reinforcement Learning Framework for LLMs

Mon, 01 Jan 0001 00:00:00 +0000

The most exciting frontier in large language model research in 2025-2026 has not been about making models bigger. It has been about making them smarter through reinforcement learning. DeepSeek-R1 demonstrated that RL training – specifically GRPO (Group Relative Policy Optimization) – can dramatically improve a model’s reasoning capabilities, enabling chain-of-thought reasoning, self-correction, and structured problem solving that rivals much larger models. ByteDance, one of the world’s largest technology companies and the creator of TikTok and Douyin, has been applying these same techniques at scale to train its own models. VeRL is the framework behind that effort.

VeRL (Voltron Reinforcement Learning) is ByteDance’s open-source reinforcement learning framework designed specifically for LLM training. It implements state-of-the-art RL algorithms including PPO (Proximal Policy Optimization) and GRPO, integrates tightly with vLLM for efficient inference during training, and supports distributed training across hundreds of GPUs. VeRL is the production framework that powers ByteDance’s internal LLM development, including the Doubao (豆包) AI assistant.