Posting peluncuran Jev
Diogo Almeida memperkenalkan Jev, model keputusan terstruktur dari TypeSafe. Tonton film peluncuran dan eksplorasi thread asli untuk pendekatan model, contoh, dan kinerja yang dilaporkan penulis.
Latest addition
Source check
Recent activityA FIELD GUIDE TO JEV / VOL. 01
Yang terbaik dari Jev: proyek nyata, panduan praktis, dan ide dari seluruh internet.
Diogo Almeida memperkenalkan Jev, model keputusan terstruktur dari TypeSafe. Tonton film peluncuran dan eksplorasi thread asli untuk pendekatan model, contoh, dan kinerja yang dilaporkan penulis.
Pengumuman peluncuran resmi TypeSafe memperkenalkan laboratorium dan mengarahkan pengembang ke Jev. Termasuk film peluncuran Diogo Almeida dan pengantar asli.
TypeSafe mengumumkan akses publik ke Jev tanpa daftar tunggu. Konsol resmi adalah titik awal untuk mencoba keputusan bertipe dalam aplikasi Anda sendiri.
TypeSafe menyoroti pendekatan Jev untuk output bertipe: aplikasi menyediakan state dan pertanyaan, lalu bertindak berdasarkan probabilitas, pilihan, atau skor. Pengumuman terkait juga mencakup integrasi API Venice.
Start with the Launch Post. Explore four official signals.
1.198 sumber
AI-assisted summaries and translations. Check original sources for context and performance claims.
Jalur terpendek dari kunci API ke keputusan yang diberi tipe.

Jev HOLDFIT 25%CAT 98%
Agens Jev yang sangat cepat untuk penggunaan browser: Jev memilih operasi dan elemen DOM dalam satu permintaan; LLM kecil hanya menulis teks saat pengetikan diperlukan.

Source-linked curation
Klasifier yang menilai kesulitan prompt secara real-time, memungkinkan mode cepat untuk prompt sederhana untuk meningkatkan interaksi pengguna.
Source-linked curation
Eksperimen menggabungkan rendering JSON dengan Jev untuk menciptakan antarmuka pengguna generatif instan untuk rendering komponen dan sistem desain yang cepat.
Source-linked curation
Plugin Claude Code yang mengganti ringkasan kompaksi dengan keputusan Jev: setiap panggilan alat dan hasilnya dinilai dalam satu permintaan cepat, yang kedaluwarsa dihapus atau dipotong, semua yang disimpan tetap dalam bentuk asli.

Source-linked curation
Posting peluncuran: arsitektur, pelatihan RLCD, harga, demo Doom dan Wikiracing, FAQ.

Jev PASSFIT 95%CAT 98%
Simulasi di mana MOSS mengangkat sampah berdasarkan instruksi, bertujuan meningkatkan kebersihan jalan melalui pengambilan keputusan otomatis.
Source-linked curation
Mode kegagalan yang diketahui dari model publik saat ini, dicatat oleh TypeSafe.
Jev HOLDFIT 78%CAT 34%
Tempelkan status, tambahkan pertanyaan, dan lihat jawaban yang diberi tipe di browser.
Jev HOLDFIT 18%CAT 41%
Jev mengontrol Super Mario Bros. menggunakan inferensi cepat dan output terstruktur, menunjukkan potensi untuk aplikasi real-time.
Source-linked curation
Ketentuan permintaan dan respons untuk POST /v1/systemone.
Jev HOLDFIT 16%CAT 100%
Klien Python resmi untuk sinkron dan asinkron. pip install typesafe-sdk.
github.comtypesafe-ai/typesafe-sdk-pythonLihat proyek ↗
Jev HOLDFIT 7%CAT 100%
Pengalihan spekulatif, pengiriman berdasarkan kepercayaan, penilaian komposit, dan pengiriman inten.
Jev HOLDFIT 10%CAT 92%
Eksperimen pencobaan pakaian virtual real-time menggunakan Jev untuk memilih dan mengganti pakaian berdasarkan masukan pengguna dan analisis transkrip.
Source-linked curation
Demo pencarian penerbangan oleh Gregor Zunic dengan ruang tindakan DOM dinamis.
Source-linked curation
Pilihan, Skor, dan Noul: tiga jenis pertanyaan dan apa yang mereka kembalikan.
Jev HOLDFIT 9%CAT 100%
Pengantar, primitif, pola, buku masak, API HTTP, dan referensi SDK.
Jev HOLDFIT 12%CAT 99%
Demo yang menunjukkan Jev memilih model AI paling sesuai untuk tugas generasi video dan gambar di Higgsfield.
Source-linked curation
Resep yang dapat direproduksi: pertanyaan paralel, reranking, pembatas, pemeriksaan kutipan, ekstraksi, klasifikasi hierarkis.
Jev HOLDFIT 14%CAT 30%
Metodologi evaluasi yang diterbitkan dan hasil per-model untuk alur kerja System One.
Jev HOLDFIT 28%CAT 95%
Argumen untuk kecerdasan yang dirancang khusus untuk perangkat lunak, bukan percakapan.
Jev HOLDFIT 13%CAT 80%
Demo interaktif resmi tentang pertanyaan berjumlah banyak dalam satu panggilan, kode menyimpan jawaban yang relevan.
Jev HOLDFIT 24%CAT 27%
A community project using Jev to control a Tesla Full Self Driving simulator, offering a quick rebuild.
Source-linked curation
Using Jev for real-time control of a robot arm in MuJoCo, with a two-step decision process for actions and movements.
Source-linked curation
Pembaruan produk dan penelitian.
x.com@typesafeai di XLihat proyek ↗
Jev HOLDFIT 10%CAT 77%
Server resmi TypeSafe. Demo Builder tersedia di saluran Show and Tell.
discord.ggDiscordLihat proyek ↗
Jev HOLDFIT 4%CAT 69%
Referensi resmi TypeSafe: Pengantar.
Jev HOLDFIT 12%CAT 99%
A community post describing a robot arm controlled by Jev, using cameras and decision-making at two decisions per second.
Source-linked curation
Pengganti TypeSafeClient resmi yang didukung oleh API LLM OpenAI, Anthropic, dan kompatibel, untuk membandingkan Jev dengan model chat.
github.comtypesafe-ai/system-one-adapter-pythonLihat proyek ↗Source-linked curation
Minimal grammY Telegram anti-spam bot powered by TypeSafe Jev.

Source-linked curation
Perencana query berbasis Jev yang meningkatkan kinerja Postgres dengan peningkatan kecepatan 12% pada benchmark urutan gabungan.
Source-linked curation
A developer project generating paintings based on Jev's pixel probability calculations, collected for the tools shelf.

Source-linked curation
Uses Jev to play Pokémon Showdown with mixed results, demonstrating its decision-making capabilities in a game context.
Source-linked curation
Utilizes Jev as a decision engine to manage and coordinate multiple bot agents for complex tasks.
Source-linked curation
A Jev-powered system that analyzes PDFs page by page to determine which require OCR processing.
Source-linked curation
Jev differs from LLMs by evaluating predefined decisions directly, enabling parallel processing of independent questions.
Source-linked curation
A tool using Jev to efficiently evaluate content by producing probabilities instead of text, faster and cheaper than traditional models.
Source-linked curation
An experiment exploring Jev's ability to solve pathfinding problems compared to traditional algorithms like A*.
Source-linked curation
A chat bot using Jev without a large language model, providing instant responses by integrating web search, Wikipedia, weather, and other tools.
Source-linked curation
Jev handles seven SEO and GEO workflows, including competitor page analysis, citation likelihood, and content optimization.
Source-linked curation
A developer project offering 110 interactive use cases, games, and scenarios with editable prompts.

Source-linked curation
A Rubik's Cube that uses Jev AI to solve itself using a human-like method rather than optimal algorithms.
Source-linked curation
Integration of Jev into a game for fast and cost-effective performance.
Source-linked curation
Browser automation where an LLM plans and Jev (Typesafe System One) decides. Library, CLI and MCP server.

Source-linked curation
Jev demonstrates real-time decision-making in the game Doom, making moves up to ten times per second.
Source-linked curation
Jev is used to identify and flag low-quality text generated by Opus.
Source-linked curation
A resource detailing a test of Jev on 12 real-world scenarios, including speed, cost, and practical applications.
Source-linked curation
Vercel's open agent framework, which ships Jev as the default evaluation model in its experimental evaluate path.

Source-linked curation
Integrates Jev into apps for smart autofill features improving user experience.
Source-linked curation
A plugin integrates Jev with a browser in Cline, enabling it to perform web-based tasks through a desktop application.
Source-linked curation
A community post describing adaptive in-app support using Jev to provide real-time assistance based on user struggles.
Source-linked curation
Classifies tax documents using an LLM pipeline for efficient processing.
Source-linked curation
A demo showing how to make open source models behave like Jev using inference engineering and scoring endpoints for decision-making.
Source-linked curation
A Chrome extension that organizes bookmarks using user-defined rules for categorization.

Source-linked curation
Jev meningkatkan efisiensi pipeline keamanan, menawarkan pengurangan biaya 5x dan pemrosesan yang lebih cepat sambil mempertahankan akurasi yang lebih tinggi dibandingkan model yang lebih kecil.

Source-linked curation
A community post describes using Jev with an LLM to play Doom, exploring multiple future outcomes.
Source-linked curation
Jev's effectiveness in performing intent-based search within Gmail.
Source-linked curation
A community post suggesting Jev can be hired as a QA tester through @typesafeai.
Source-linked curation
A community project exploring Jev's potential to evaluate agent progress beyond basic trace data.
Source-linked curation
An experiment to enable Jev to perform dogfighting, with ongoing improvements.
Source-linked curation
A community post demonstrating Jev playing Flappy Apex, achieving a high score autonomously through Appduct and Fable.
Source-linked curation
A trading bot using Jev to decide buy/sell actions based on asset price feeds and executing real trades.
Source-linked curation
A community post compares Jev's performance to Astra in gaming scenarios, highlighting its unique approach.
Source-linked curation
A GitHub project converting agent skills into Oxlint plugins that involve Jev.

Source-linked curation
A setup using Jev and GrokBot to route AI agent tasks efficiently, reducing costs and execution time with a seven-step configuration process.
Source-linked curation
A 30-minute internal session on Jev attended by over 50 engineers, generating more than 50 product ideas.

Source-linked curation
A task using Jev to score 3,000 kids' snacks with multiple criteria in 28 seconds.
Source-linked curation
A game where Jev makes decisions to keep a city on a whale alive, integrated with GPT-6 Astra and H3 Max.
Source-linked curation
Moved hard-coded rules to Jev for faster AI processing and reduced costs in projects like aiseotracker and linkdr.
Source-linked curation
A developer project providing a local MCP server for running pre-made Jev question packs.

Source-linked curation
A podcast segment discussing Jevons paradox and its connection to TypeSafe AI's funding.
Source-linked curation
Non-autoregressive decision engine on ModernBERT (151M) with calibrated uncertainty (RLCD), TypeSafe AI Jev benchmark audit, and in-browser WebGPU playground.

Source-linked curation