Will this generate code or leak my key into logs?

It only ranks words already in the buffer and never generates code. Keys come from the environment and are never logged; without a key it emits one warning and stays in fuzzy-only mode.

Why did the menu appear but never reorder?

Set debug_guards to true and the log names the guard that rejected the update. Also confirm the key is visible to Neovim and that completefunc reports v:lua.JevComplete with an allowed filetype.

How do I keep it from firing on every keystroke?

Leave auto_trigger off and fire manually with C-x C-u. Opting in waits a longer debounce after your last keystroke, skips readonly buffers and disabled filetypes, and a word-ending character cancels a pending trigger.

Does it work in huge files?

Extraction caps at max_extract_words so later words in huge buffers are skipped, with context falling back to lines around the cursor over the token ceiling. The benchmark script reports timings for tuning both.

What do I need to install and run it?

Neovim 0.12 or newer, a TypeSafe API key, and curl in PATH. Install with lazy.nvim pointing at tibuchivn/jev-complete.nvim and calling require("jev").setup(), or clone the repository, prepend it to runtimepath and call setup().

What are the key configuration options?

confidence_threshold (0.7, below which the fuzzy order stands), max_candidates (30), jev_question_format (noul, choice or score, with noul the cheapest default), model jev-latest, debounce_ms (300 manual and 500 auto), max_context_tokens (28000), plus disabled_filetypes, auto_trigger and menu_update_mode.

How does the async reordering stay safe?

The completefunc cannot return asynchronously, so the fuzzy list appears immediately and is replaced once Jev answers. Three guards protect the update: the popup must still be visible, the cursor must not have moved, and the typed prefix must be unchanged. menu_update_mode chooses between an in-place update and re-triggering completion.

CommunityJev Tools & CLIs

Jev Complete.nvim

Neovim plugin that uses Jev as a ranking oracle: buffer words appear instantly as fuzzy matches, then Jev re-ranks up to 30 candidates against surrounding code and rebuilds the menu.

See highlights

Open github.com

Stars: 1
Language: Lua
License: MIT
Published by: Community
Author: tibuchivn
Last verified: Oct 2026

tibuchivn/jev-complete.nvim

User repository on GitHub

View on GitHub

Semantic code completion for Neovim using Jev (TypeSafe AI) as a ranking oracle. Fuzzy candidates appear instantly; Jev re-ranks them by code context.

Forks: 0
Last pushed: Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.

About this project

Semantic code completion for Neovim that uses Jev (TypeSafe AI) as a ranking oracle. Candidates come from the current buffer and appear instantly as fuzzy matches; Jev then re-ranks them against the surrounding code and the menu is rebuilt in the better order. Jev ranks. It does not generate code. Clone the repository, add it to runtimepath, and call setup(): Add that line to your shell profile to make it persistent. The legacy variable TYPESAFEAPIKEY still works, and vim.g.jevapikey takes precedence over both:

Quoted from the project's README. Read the full README.

Tagscommunity github

Works forJev for Ranking

Used inJev for Software & DevTools

Highlights

Jev ranks existing buffer words only; without a key completion stays fuzzy-only with a warning.
Manual trigger shows fuzzy candidates at once, then rebuilds the menu in Jev order after about a second.
Reorders only above a 0.7 confidence threshold, with noul, choice, or score formats and a 3s timeout.
Guards require a visible popup, unmoved cursor, and unchanged prefix, with flags logging rejections.

Watch out

Needs Neovim 0.12 or newer, curl, and a key via JEV_API_KEY, TYPESAFE_API_KEY, or vim.g.jev_api_key; Jev latency is commonly 0.5-1.5s and auto-trigger increases API calls. Copy reviewed against the repository README on 2026-10-04, not from running it.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

paper-radar-jev

Python CLI that pulls recent arXiv papers, scores relevance with TypeSafe and writes daily Markdown and JSON rankings with resume-safe storage.

Jev Tools & CLIs#community#github

Community

LYchoon/paper-radar-jevPython

GenOffice

9.2k

An open-source AI office suite whose local file search reranks the top 20 FTS hits with TypeSafe Jev relevance scores before showing results.

Jev Tools & CLIs#community#typescript#desktop

Community

genspark-ai/genofficeTypeScript

MemSearch

2.7k

A cross-platform memory layer for AI coding agents: conversations from Claude Code, Codex, DeepSeek Harness, OpenClaw, and OpenCode become Markdown memories indexed in Milvus, with optional Jev reranking through the TypeSafe API.

Jev Tools & CLIs#community#python#agents

Community

zilliztech/memsearchPython

XERJ

3.5k

An Elasticsearch-compatible autoindex search layer that optionally sends top hits to TypeSafe Jev and reorders results by relevance scores.

Jev Tools & CLIs#community#rust#search

Community

xerj-org/xerjRust

Beacon

1.8k

An open-source, local-first memory layer that captures session history across Claude Code, Cursor, Codex, OpenCode, Cline, and 20+ other harnesses, then has Jev judge bounded, redacted trace projections for reusable lessons a person reviews.

Jev Tools & CLIs#community#go#agents

Community

Asymptote-Labs/agent-beaconGo

jevgrep (code search)

2.6k

A CLI that asks Jev which files and declarations answer a repository question, then prints verbatim excerpts for a coding agent to use.

Jev Tools & CLIs#community#typescript#cli

Community

dzhng/jevgrepTypeScript

From the Jev guides

Original write-ups on what this entry is for.

Jev on the Gateways: Vercel, Workers AI, and OpenRouter

Oct 12, 2026

The request shape barely changes across surfaces; the surrounding contract does. When to call the API directly and when a gateway, a Worker binding, or a router entry earns its extra hop.

Jev Tools & CLIs#gateway#vercel#cloudflare

Read article

Exporting Decision Models: GGUF, ONNX, and Device Runtimes

Oct 10, 2026

Taking a decision checkpoint out of Python: what the GGUF and browser WebGPU exports preserve, what they change, and the parity evidence the model cards publish.

Jev Tools & CLIs#open-models#local#onnx

Read article

Records at Rest: Documents, Inboxes, and Risk Queues

Sep 28, 2026

Jev labels a record that already exists. Code moves, flags, or holds it. How the document, mail, and transfer tools keep the action out of the model.

Jev Tools & CLIs#classification#email#security

Read article

Back to all resources

From the Jev community

Posts from builders shipping with Jev right now.

Follow @typesafeai