JevTypeSafe AI
|
Mercury 2.5Inception
|
|||||
Related Products
|
||||||
About
Jev is a System One Model from TypeSafe AI designed for fast, structured decision-making inside software and automated workflows. Unlike conventional large language models that generate free-form text token by token, Jev accepts unstructured state and returns predefined type-safe values with calibrated probabilities and confidence scores. The model uses a parallel sampling architecture and a training approach called Reinforcement Learning for Calibrated Decisions, which is optimized for producing reliable probabilistic decisions rather than conversational responses. Jev is intended for tasks such as classification, routing, scoring, extraction, branching logic, verification, guardrails, and other AI-powered workflows where structured outputs can plug directly into application code. TypeSafe reports end-to-end response times ranging from about 70 to 500 milliseconds and says Jev can be substantially faster and more efficient than frontier language models on System One-style workloads.
|
About
Mercury 2.5 is Inception’s most capable production model yet and a significant step up in quality over Mercury 2 while maintaining the same low-latency serving profile. It is the most capable diffusion LLM on the market and, according to Inception, the largest diffusion language model ever trained. Mercury 2.5 delivers a 40% increase in intelligence over Mercury 2, with performance comparable to cost-optimized frontier models such as GPT-5.6 Luna (Low), Gemini 3.5 Flash-Lite, and Claude Haiku 4.5. It generates at 1,107 tokens per second on widely available NVIDIA GPUs and supports a 260K-token context window. Capabilities include tunable reasoning, parallel tool calls, and schema-aligned JSON. The model is designed for latency-sensitive workloads where many model calls may happen inside a single interaction. In search agents and RAG pipelines, it can support planning, query rewriting, reranking, fact structuring, source summarization, and answer checking while keeping calls fast.
|
|||||
Platforms Supported
Windows
Mac
Linux
Cloud
On-Premises
iPhone
iPad
Android
Chromebook
|
Platforms Supported
Windows
Mac
Linux
Cloud
On-Premises
iPhone
iPad
Android
Chromebook
|
|||||
Audience
Software developers, AI engineers, automation teams, infrastructure builders, and enterprises that need low-latency, type-safe, probabilistic AI decisions for production software and structured workflows
|
Audience
Developers and enterprises seeking a high-speed diffusion language model for search, RAG, voice agents, coding assistants, and latency-sensitive AI applications
|
|||||
Support
Phone Support
24/7 Live Support
Online
|
Support
Phone Support
24/7 Live Support
Online
|
|||||
API
Offers API
|
API
Offers API
|
|||||
Screenshots and Videos |
Screenshots and Videos |
|||||
Pricing
Input: $0.042 / 1M tokens
Input tokens: $0.042 / 1 million tokens ($42 per billion tokens).
Output tokens: FREE (too cheap to meter).
Free Version
Free Trial
|
Pricing
No information available.
Free Version
Free Trial
|
|||||
Reviews/
|
Reviews/
|
|||||
Training
Documentation
Webinars
Live Online
In Person
|
Training
Documentation
Webinars
Live Online
In Person
|
|||||
Company InformationTypeSafe AI
Founded: 2024
United States
typesafe.ai/
|
Company InformationInception
United States
www.inceptionlabs.ai/blog/introducing-mercury-2-5
|
|||||
Alternatives |
Alternatives |
|||||
|
|
|
|||||
|
|
|
|||||
|
|
||||||
Categories |
Categories |
|||||
Integrations
JSON
|
||||||
|
|
|