Chrome Extension for Agentic Browsing
This Chrome extension allows agents to browse, click, and interact with websites directly in the browser. It utilizes Jev by typesafeai and includes AI Gateway by Vercel.
view details →JEV use cases
94 published use cases that name JEV.
Model Benchmark: Jev vs. Claude for Web Forms
A benchmark was conducted comparing Jev by TypeSafe and Claude Code agents for filling out web forms. The test involved three real forms, with no submissions made.
view details →Jev Adapter for Prompt Classification
The Jev adapter classifies prompts before they reach the LLM, reducing token consumption and costs. It provides classification and confidence scores, marking prompts for review when confidence is near the threshold.
view details →Tiny Support Ticket Classifier Using JEV
This support ticket classifier built with JEV categorizes tickets, assesses urgency, and determines if human intervention is needed. It has achieved 100% accuracy in initial tests and operates at a lower cost.
view details →Fraud Detection Pipeline Using Jev and Kimi K3
Jev classified 100 emails in 1.42 seconds, achieving 96% accuracy by routing uncertain cases to Kimi K3. The total inference cost was approximately $0.07.
view details →Evaluation of Jev Model for SEC Filings and Fact Extraction
The Jev model was tested on SEC filings and fact extraction tasks, showing strengths in narrow yes-or-no questions but weaknesses in understanding document context. It performed well as a first-pass filter but lagged behind production models in accuracy.
view details →Content-Aware PII Redaction in Postgres
This system uses Jev to tag personal information in support messages, allowing selective redaction. The redact() SQL function ensures that only sensitive data is masked while keeping the rest of the message readable.
view details →Easy-Jev Demo for Real-Time Classifications
Easy-Jev is a demo that allows users to change inputs and see real-time classification changes. Developed by a master's thesis researcher in classification models, it showcases a modern approach to combining different models for problem-solving.
view details →Realtime Adblocker Extension with Typesafe
This extension classifies DOM elements as ads or non-ads in real-time and removes ads accordingly. It represents a significant advancement in the application of AI technology.
view details →Ask Jev Anything: A Unique Q&A Experience
Engage with Jev by asking questions at the provided link. This interactive experience focuses on judgment rather than direct answers.
view details →Jev Decision Model Performance Comparison
Jev, a new decision model, demonstrated a routing cost 125 times lower than GPT-5.5 when deciding actions for a coding agent after a failed test.
view details →Impact of New Tool on Code Review Process
A new tool introduced at the company has already influenced code review practices within 24 hours of its launch. A guide based on internal metrics is available for reference.
view details →Expense Categorization Using Jev
The author utilizes Jev to categorize company expenses in Mongolian, leveraging a year’s worth of Excel files for training. This approach allows for unique classification tailored to their specific needs.
view details →Listing by Greg PR
Explore the latest updates and insights shared by Greg PR on X.
view details →Cost-effective AI agent society
A society of 100 agents was built using the Jev model, enabling 100 decisions per request in approximately 250ms. This setup allows for 20,000 decisions at a cost of just 11 cents, making it 240 times cheaper than a frontier model.
view details →Trading Bot Built with Jev
This trading bot determines whether to buy or sell based on asset pair price feeds and executes real trades. It utilizes Monad to place orders on Kuru's on-chain order book every 300ms.
view details →Hermes Setup with 443 Unique Skills
The Hermes setup now features 443 unique skills, utilizing Jev for skill selection. Jev achieved 94.4% exploratory reused-data performance, ensuring confidence-aware decision-making.
view details →Repository Link Shared by User
User shared a repository link in response to requests. Access the repo for more information.
view details →Jev Model Evaluation Results
The Jev model from Typesafe AI was tested on Modemdev's evaluation for agent replies, showing an 8x speed improvement and a 25x cost reduction.
view details →Integration of Jev for Decision-Making at The Grand Internet Hotel
The Grand Internet Hotel has integrated Jev to enhance decision-making processes for agents. It evaluates structured questions and provides decisions with confidence estimates, ensuring actions align with hotel rules.
view details →Jev: Fast Decision-Making for LLMs
Jev offers a new intelligent decision-making primitive that enhances model routing and classification tasks. It allows for quick and cost-effective decision-making, improving the efficiency of LLM calls in applications.
view details →Testing New Foundation Model with Impressive Results
A new foundation model has been tested for about a week, showing potential to be indispensable in the next 6-12 months. It operates by producing probabilities instead of words, demonstrating efficiency with 25x faster processing and 600x lower costs compared to traditional models.
view details →