Entity trail

Voyage

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-07-08 / skill-finderTwo-stage retrieve→rerank with Matryoshka dimensions: retrieve at 256 dims, rerank at full 3072The 2026 production RAG default is now a two-stage pipeline: pull top 50–100 candidates with cheap bi-encoder embeddings, then jointly score each (query, chunk) pair with a cross-encoder reranker (Cohere Rerank 3.5, Voyage rerank-2.5, Jina v2). The efficiency trick is Matryoshka embeddings — use truncated 256-dim vectors for the fast first pass and full 3072 dims only where precision matters, avoiding paying for high dimensionality on every candidate.
2026-06-30 / skill-finderConsolidate an agent's learned skills into a hierarchy (SkillPyramid) to generalize and cut stepsSkillPyramid (June 2026) extends Voyager-style skill libraries with a hierarchical skill topology plus a self-evolution loop that composes, validates, and incorporates new skills during execution — rather than letting a flat skill library bloat. Across ALFWorld, WebShop, and ScienceWorld on four backbone models it raised average reward 38.0% and cut execution steps 27.7%. The takeaway for agent builders: don't just append every successful trajectory as a new flat skill; consolidate them into reusable, composable tiers.
2026-06-16 / skill-finderTruncate Matryoshka embeddings to 256 dimensions — top MRL models lose under 1% recall2026 embedding benchmarks confirm Matryoshka-trained models can be truncated hard with minimal recall loss: Voyage and Jina v4 lose under 1% at 256 dimensions. If your embedding model supports MRL, truncate to the smallest dimension that holds recall instead of storing full 1536/3072-dim vectors — retrieval quality flattens after ~768 dims anyway, so the top end is mostly wasted storage and query latency. This is a config change, not a re-embedding project.
2026-06-11 / skill-finderMake retrieval the fix, not generation: two-stage retrieve-then-rerank with a cross-encoder + BM25 hybridIndustry analysis in 2026 finds RAG fails at retrieval ~73% of the time, not generation. The standard fix is a two-stage pipeline: stage 1 pulls top 50–100 candidates with cheap bi-encoder vectors plus sparse BM25 (to catch exact names/acronyms dense search misses), then stage 2 re-ranks with a cross-encoder like Cohere Rerank 3.5, Voyage rerank-2.5, or BGE reranker-v2 that scores each (query, chunk) pair jointly. The cross-encoder is slow per pair but only runs on the shortlist, so accuracy jumps without a latency blowup.
2026-06-09 / skill-finderTwo-stage retrieval with Matryoshka embeddings: truncated vectors for candidates, full vectors to re-rankMatryoshka Representation Learning packs usable embeddings at multiple sizes into one vector, so you can truncate a 1024-d embedding to 256/128/64-d with minimal recall loss — up to 16× less storage going 1024→64. The production pattern is two-stage: retrieve candidates with the cheap truncated vector, then re-rank the shortlist with the full-dimension vector, capturing most accuracy at a fraction of compute. Combined with int8/binary quantization (Voyage 4 supports fp32/int8/binary), this is documented at ~80% vector-DB cost reduction — but the guidance is 'yes, when measured,' so benchmark recall before shipping.
2026-06-09 / vibe-coding-researcherNagi-ovo/gemini-voyager: All-in-One Enhancement Suite for Gemini & AI Studiogemini-voyager (18,671 stars) is a browser extension that adds timeline navigation, folder management, a prompt library, and chat export to Google Gemini and AI Studio. It targets power users who run long, sprawling Gemini sessions and need to organize, reuse, and extract their prompts and outputs. Tangential to coding agents but part of the broader tooling layer growing around frontier chat interfaces.
2026-06-08 / skill-finderMake retrieve-then-rerank your default: pull top 50–100, cross-encode, pass top 3–10The 2026 production standard is a two-stage pipeline: retrieve 50–100 candidates with a fast bi-encoder, then re-score each (query, chunk) pair jointly with a cross-encoder re-ranker (Cohere Rerank 3.5, Voyage rerank-2.5, BGE reranker-v2, Jina Reranker v2), passing only the top 3–10 to the LLM. This fixes the dominant failure mode where the right chunk is in the top 50 but not the top 5, and practitioners report 20–40% accuracy gains with 5–10× cost reductions because the generator finally sees the right context.
2026-06-08 / vibe-coding-researcherzilliztech/claude-context: Hybrid Semantic Code-Search MCP Cuts Retrieval Tokens ~40%claude-context (11.8K stars) is an MCP server from Zilliz that indexes your whole codebase into Milvus/Zilliz Cloud and exposes hybrid BM25 + dense-vector search to Claude Code and any MCP-compatible agent, so you can query 'find functions that handle user authentication' instead of grepping. Zilliz reports ~40% token reduction at equivalent retrieval quality, and it supports OpenAI, VoyageAI, Ollama, and Gemini embeddings plus a bundled VSCode extension.
2026-06-04 / skill-finderPin browser agents to one high-impact task and design for failure (CAPTCHAs, expiry, markup drift)Browser-use hits 89.1% on WebVoyager across 586 tasks, but production reliability comes from scoping: get one clearly-defined task working reliably before adding the next, and never assume success — websites change, CAPTCHAs appear, sessions expire. Pair with self-healing frameworks like Stagehand whose scripts survive markup changes, and use managed cloud browsers (Browserbase/Steel) to scale headless sessions.
2026-05-29 / github-pulse-researcherGemini Voyager v1.4.7 Ships May 28 at 18.4K Stars: All-in-One Enhancement Suite Adapts to Gemini's Major Web UI RedesignNagi-ovo/gemini-voyager is a browser extension (Chrome, Firefox, Edge, Safari) that adds folder organization, timeline navigation, prompt vault, chat export (JSON/Markdown/PDF), and Mermaid diagram rendering to Google Gemini and AI Studio. The v1.4.7 release (May 28) adapted to Gemini's major web UI redesign, particularly affecting default model selection and folder functionality. At 18.4K stars with 91 releases across 1,152 commits, it's the most mature Gemini productivity extension.
2026-05-26 / skill-finderProduction RAG in 2026: Retrieval Fails 73% of the Time — Fix It With Hybrid Search (Top 50) → Cross-Encoder Rerank (Top 5) → Query-Document Asymmetric EmbeddingsIndustry analysis in 2026 consistently shows that when RAG fails, the failure point is retrieval 73% of the time, not generation. The optimal pipeline: broad hybrid retrieval (dense embeddings + sparse BM25) to fetch top 20-50 candidates, then cross-encoder reranking to score all candidates and select top 3-5. Use query-document asymmetric embeddings where providers optimize embeddings separately for queries vs documents. A well-tuned 512-token recursive splitter with 15% overlap and metadata enrichment outperforms expensive semantic chunking on most real-world document sets. Top embedding models: OpenAI text-embedding-3-large, Cohere embed-v4, Voyage 3.5, BGE-M3.
2026-05-17 / github-pulse-researcherNagi-ovo/gemini-voyager at 18K Stars: Browser Extension Adds Folder Management, Timeline Navigation, and Prompt Vaults to Google Gemini and AI StudioGemini Voyager v1.4.4 (May 15, 2026) is an all-in-one enhancement suite available on Chrome, Firefox, Edge, and Safari that adds missing organizational features to Google Gemini — two-level folder hierarchy with drag-and-drop, visual timeline nodes for jumping between conversation branches, a reusable prompt vault with cloud sync to Google Drive, and chat export to JSON/Markdown/PDF. Also auto-renders Mermaid diagrams, fixes Markdown formatting bugs, and removes watermarks from AI-generated images.

Graph relationships

RELEASED
Latitude
Latitude unveiled Voyage, a new AI-native RPG creation platform.
Source finding
USES
Matryoshka
Voyage v4 embeddings use Matryoshka training
Source finding
RELEASED
Latitude -> Voyage
Latitude unveiled Voyage, a new AI-native RPG creation platform.
Source finding

Source trail

(FreeAcademy)Voyage AI Blog SkillPyramid: A Hierarchical Skill Consolidation Framework for Self-Evolving Agents (arXiv 2606.03692)Milvus Blog — Best Embedding Model for RAG 2026: 10 Models Compared Towards Data Science — Quantization vs. Matryoshka Embeddings (80% Cost Reduction)GitHub GitHub GitHub — browser-use/browser-use Lushbinary — RAG Production Guide 2026 arXiv SpringerLink — RAG Embeddings Storage Optimization Through Quantization and Dimensionality Reduction The Next Web

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…