Research Papers 论文研究 4h ago Updated 31m ago 更新于 31分钟前 47

Giga-Embeddings: Mixture-of-Experts Encoders for High-Throughput Text Embeddings Giga-Embeddings:用于高吞吐量文本嵌入的混合专家编码器

Giga-Embeddings introduces a sparse 10B-parameter Mixture-of-Experts encoder with ~1.8B active parameters per token, achieving top retrieval quality across English, Russian, multilingual, and code benchmarks The model processes 114.5k tokens/second with 1024-token inputs in vLLM, delivering 25% higher throughput than a dense 3B model and 1.56–2.65x the throughput of evaluated external systems The family includes a dense 3B encoder and a distilled 480M compact model trained with a dimension-agnos Giga-Embeddings是文本嵌入模型家族,最大成员为稀疏10B参数MoE编码器,每token约1.8B活跃参数 在英语、俄语、多语言和代码MTEB基准测试中均取得家族内最强性能 vLLM基准测试中处理114.5k tokens/秒,吞吐量比密集3B模型高25%,比外部系统高1.56-2.65倍 模型家族还包括密集3B编码器和蒸馏480M编码器,后者在俄罗斯MTEB上得分70.98,超越FRIDA但参数少42% 所有三个模型检查点已开源发布

62
Hot 热度
72
Quality 质量
68
Impact 影响力

Analysis 深度分析

TL;DR

  • Giga-Embeddings introduces a sparse 10B-parameter Mixture-of-Experts encoder with ~1.8B active parameters per token, achieving top retrieval quality across English, Russian, multilingual, and code benchmarks
  • The model processes 114.5k tokens/second with 1024-token inputs in vLLM, delivering 25% higher throughput than a dense 3B model and 1.56–2.65x the throughput of evaluated external systems
  • The family includes a dense 3B encoder and a distilled 480M compact model trained with a dimension-agnostic objective that aligns teacher-student similarity distributions
  • The 480M model scores 70.98 on Russian MTEB, surpassing FRIDA while using 42% fewer parameters
  • All three model checkpoints are publicly released

Why It Matters

This work demonstrates that sparse Mixture-of-Experts architectures can break the traditional tradeoff between embedding quality and serving throughput, which is critical for production RAG and retrieval systems. The release of a compact distilled variant that outperforms larger competitors with significantly fewer parameters provides a practical path for resource-constrained deployments.

Technical Details

  • Sparse 10B MoE encoder: ~1.8B active parameters per token, achieving the strongest aggregate performance across English, Russian, multilingual, and code MTEB benchmark suites
  • Throughput performance: 114.5k tokens/second at 1024-token input length using vLLM, outperforming a dense 3B model by 25% and external systems by 1.56–2.65x
  • Model family: Three checkpoints — sparse 10B MoE, dense 3B, and distilled 480M — covering a range of compute and memory budgets
  • Distillation method: A dimension-agnostic objective that aligns teacher and student similarity distributions, enabling the 480M model to achieve 70.98 on Russian MTEB with 42% fewer parameters than FRIDA
  • Benchmarks: Evaluated across four MTEB suites (English, Russian, multilingual, code)

Industry Insight

  • Sparse MoE encoders are becoming a viable production strategy for high-throughput embedding serving, offering a compelling alternative to dense models that scale linearly in both parameters and inference cost
  • The dimension-agnostic distillation objective presents a reusable technique for compressing large embedding models without sacrificing retrieval quality, particularly valuable for edge and cost-sensitive deployments
  • With all checkpoints released, practitioners can adopt a tiered strategy: the 10B MoE for maximum retrieval accuracy, the 3B dense model as a balanced middle ground, and the 480M model for low-latency, resource-constrained environments

TL;DR

  • Giga-Embeddings是文本嵌入模型家族,最大成员为稀疏10B参数MoE编码器,每token约1.8B活跃参数
  • 在英语、俄语、多语言和代码MTEB基准测试中均取得家族内最强性能
  • vLLM基准测试中处理114.5k tokens/秒,吞吐量比密集3B模型高25%,比外部系统高1.56-2.65倍
  • 模型家族还包括密集3B编码器和蒸馏480M编码器,后者在俄罗斯MTEB上得分70.98,超越FRIDA但参数少42%
  • 所有三个模型检查点已开源发布

为什么值得看

本文展示了MoE架构在文本嵌入任务中的高效应用,为高吞吐检索场景提供了兼顾质量与性能的新方案。蒸馏技术的创新目标函数设计为小模型压缩提供了新思路,对工业界部署有直接参考价值。

技术解析

  • 核心架构为稀疏10B参数Mixture-of-Experts编码器,每token仅激活约1.8B参数,在保持强检索质量的同时实现高效推理
  • 模型家族覆盖三个规模:10B MoE、3B密集和480M蒸馏版本,通过维度无关的目标函数对齐教师和学生相似度分布进行蒸馏训练
  • 在vLLM基准测试(1024-token输入)中达到114.5k tokens/秒的处理速度,比密集3B模型高25%,比外部系统高1.56-2.65倍
  • 480M蒸馏模型在俄罗斯MTEB上获得70.98分,超越FRIDA模型但参数减少42%,验证了蒸馏策略的有效性

行业启示

  • MoE架构在文本嵌入任务中展现出显著优势,通过稀疏激活平衡检索质量与推理效率,为大规模部署提供可行路径
  • 蒸馏技术结合相似度分布对齐目标函数,可在大幅减少参数量的同时保持甚至超越更大模型性能,适合资源受限场景
  • 开源完整模型检查点降低了行业采用门槛,有助于推动文本嵌入技术的标准化和生态建设

Disclaimer: The above content is generated by AI and is for reference only. 免责声明:以上内容由 AI 生成,仅供参考。

Embedding Model 嵌入模型 Open Source 开源 Research 科学研究 Benchmark 基准测试 Inference 推理