Ultrasafe AI Model Card

What it is

Ultrasafe Mini 1.0 is an agentic language model designed for global users and organizations. It uses a dense model along with ML models and specialized variants of that model for domain-specific tasks (specialized experts for agents and tool calling, reasoning, code generation, retrieval-augmented QA, structured output, etc.) plus an integrated router that dynamically selects the best experts for each user request. This design aims to deliver fast responses on simple tasks and deeper, deliberative reasoning on harder problems—without disclosing parameter count.

Why it matters globally

The model prioritizes multilingual support (major world languages and widely used dialects), code-switched English, and mixed-script input. It is tuned for global use cases—public services, BFSI, retail, SMB enablement—and engineered to interoperate with key digital public infrastructure systems worldwide (e.g., identity frameworks, payment systems, consent frameworks, and commerce standards), while conforming to international data protection and AI governance norms (GDPR, DPDP, OECD AI principles).

What's new

3.1 Model Architecture

Core Model Design

Ultrasafe Mini 1.0 is built on a dense Transformer architecture with the following specifications:

• Dense Transformers — optimized for consistent accuracy across domains.
• SiLU activation for smoother gradient flow and improved convergence.
• 80 Transformer layers for deep contextual reasoning.
• ~155K vocabulary size covering code, Indian languages, and global languages.
• Context window: Up to 128K tokens (with 32K tokens as the optimal operational range for best performance).
• Training sequence length: Pretraining on 4K and 8K sequences, extended to 128K during post-training via RoPE extrapolation techniques.

Training Scale

• Training corpus size: 18.5 trillion tokens total.
• 4 trillion tokens — Code from 90+ programming languages.
• 14+ trillion tokens — English and other major global languages.
• Model size: Up to 40 billion parameters.
• Distilled versions: Optimized lightweight variants of larger models for low-resource deployment while retaining reasoning capability.

3.2 Datasets

Pretraining Datasets

• Primary Source: Public, high-quality datasets collected from open repositories.
• Filtering: Multi-stage filtering pipeline to retain only the highest-quality data.
• Sensitive Data Removal: Implemented advanced PII detection and removal to ensure safety and compliance before training.

Domain-Specific Collections

• Code: Curated datasets from multiple organizations and open-source projects across 90+ programming languages.
• Mathematics: Specialized datasets with step-by-step solutions for numeric reasoning.
• Reasoning: Logic, problem-solving, and multi-step reasoning datasets from diverse domains.

Synthetic Data Generation

Generated large volumes of synthetic datasets using primitive approaches tuned for:

• Indian language fluency and dialect diversity.
• Complex reasoning workflows across multiple domains.

3.3 Post-Training

Instruction-Tuning

Curated highly complex, multi-turn, and high-quality datasets covering a wide range of tasks.

Over 50+ million samples focusing on:

• Conversational flow.
• Indian language style, dialect, and vocabulary adaptation.
• Agentic workflows and multi-tool calling.
• Advanced reasoning and decision-making.

Reinforcement Learning

Stage 1 — GRPO (Guided Reinforcement Preference Optimization): Combined rule-based reward functions and LLM-based reward models. Categories included mathematics, code generation, reasoning, and agentic workflows.

Stage 2 — DPO (Direct Preference Optimization): Trained on millions of preference pairs to better align with user conversational styles and cultural expectations.

3.4 Expert-Model Core (Dense Transformer with Integrated Router)

Ultrasafe Mini 1.0 is an agentic language model designed specifically for Indian users and organizations. It uses a dense Transformer architecture combined with auxiliary ML models and specialized variants of the base model for domain-specific tasks—such as agent and tool calling, reasoning, code generation, retrieval-augmented QA, and structured output generation.

An integrated routing system dynamically selects the most relevant specialized variant (expert) for each request. This routing blends:

• Request-level pre-routing to set defaults for tools, safety posture, and reasoning depth before generation.
• Token- or span-level adaptation for fine-grained control over formatting, numeracy, code handling, and multilingual script diversity.

Expert specializations (illustrative):

• Multilingual/Indic variant tuned for script diversity (Hindi, Hinglish, Bengali–Assamese, Gujarati, Odia, Telugu, Kannada, Malayalam, Tamil, etc.) and code-switched Indian English.
• Retrieval/RAG variant optimized for grounding, citation formatting, and hallucination reduction.
• Code/SQL variant with constrained decoding utilities and static-analysis hints.
• Structure/formatting variant for JSON schema generation.

3.5 Router design

Inputs. The router consumes the user query, system/policy state, domain specific experts GenAI and ML models, org context (rag, database, files), task and conversation context, available tools and resources.

Decision Process: Based on the conversation and the task at hand, the router determines the optimal execution path to complete the request. It dynamically selects the most suitable expert model or tool for each subtask.

If the task is complex and spans multiple domains, the router can activate multiple experts in parallel or sequence, enabling them to collaboratively solve the problem. This allows for both specialized precision and cross-domain reasoning, ensuring that each component of the solution is handled by the most capable resource.

3.6 “Thinking” Budget & Depth Control

In the thinking model, the system can define a budget and allocate computational resources for reasoning based on task complexity. By default, Ultrasafe Mini X1 supports three reasoning depth profiles:

• Light — Minimal deliberation for straightforward tasks, prioritizes speed and low resource usage.
• Medium — Balanced reasoning for moderately complex queries; blends efficiency with accuracy.
• Complex — Deep, multi-step reasoning for high-stakes or multi-domain problems; allocates maximum resources and extended context processing.

The router automatically selects the appropriate depth profile based on task complexity, user priority, and policy constraints, but it can also be overridden by explicit user or system instructions.

3.7 Agent Framework

The Ultrasafe Mini 1.0 agent framework orchestrates sequential workflows with dynamic planning, agent/tool selection, and iterative evaluation for subsequent steps.

• The framework and model can coordinate millions of agents within a single request without requiring custom rules or hand-crafted workflows for each scenario.
• This design enables handling of diverse task types and tool integrations by default, leveraging the reasoning capabilities of the underlying models.
• Agents can dynamically collaborate, share intermediate outputs, and adapt their plans mid-execution to achieve optimal results.

3.8 Performance & Efficiency

Ultrasafe Mini 1.0 is optimized for fast, accurate responses while keeping compute usage minimal.

• KV-Chaining: Reuses key–value states across multi-step reasoning and iterative calls, reducing recomputation and latency while preserving context fidelity.
• Speculative Decoding: Produces preliminary outputs using lightweight speculative passes, verified within the same SLM to accelerate generation without compromising accuracy.
• SLM-Only Integration Flow: All tasks are handled within a Small Language Model (SLM) framework, with optimized routing and internal specialization to maintain accuracy while minimizing cost and latency—no larger models are required.

4.1 Enterprise scenarios (global)

4.2 Developers & startups

4.3 Public‑facing experiences

4.4 Contact center & operations

• Agent assist: live suggestion, disposition drafts, and compliance guardrails during calls/chats.
• Call summarization with action items, follow‑ups, and structured CRM updates.
• Quality monitoring: scorecards, coaching insights, and trend detection across languages.
• Real-time knowledge retrieval with snippet insertion.
• Auto-disposition coding and wrap-up time reduction.
• Escalation triggers with policy citations and customer history.

4.5 Data & analytics

• Retrieval‑augmented generation (RAG) with citations and source filtering.
• Text‑to‑SQL/DSL over data warehouses and lakes with safety checks.
• Semantic search, summarization at scale, and content classification/redaction pipelines.
• Document lineage, provenance tracking, and governance tagging.
• Batch pipelines for large-scale summarization, labeling, and redaction.
• BI copilots that explain anomalies and forecast drivers.

4.6 Security, risk & compliance

• PII redaction/anonymization, policy enforcement, and ethical guardrails across workflows.
• Evidence trails: signed prompts/responses, chain‑of‑custody for high‑assurance contexts.
• Content safety filters with multilingual coverage and configurable risk tiers.
• Red-team prompt libraries and continuous adversarial testing.
• Policy-as-code enforcement with explainable deny reasons.
• Data residency routing and KMS-backed encryption posture.

4.7 Insurance

• Policy question answering and coverage explainers across P&C, life, and health.
• Claims intake assistants with document checklists and fraud-risk hints.
• Underwriting support: risk factor extraction and scoring inputs from submissions.
• Regulatory form drafting and compliance summaries (e.g., Solvency II).
• Agent/broker sales aids: quote comparisons and next-best action suggestions.
• Loss run and incident report summarization with structured outputs.

4.8 Real Estate

• Property listing normalization and amenity extraction from unstructured text.
• Lease abstraction: key clause extraction and renewal/notice reminders.
• Tenant onboarding assistants and maintenance request triage.
• Market comp summaries and CMA drafting support with citations.
• Document package assembly: disclosures, addenda, and checklists.
• Multilingual community and HOA policy explainers.

4.9 Pharma

• SOP and GxP documentation assistants with audit trails.
• Clinical trial protocol summarization and eligibility screening aides.
• Adverse event intake and MedDRA coding suggestions.
• Regulatory submission checklist generation (e.g., eCTD).
• Medical information response drafting with reference citations.
• Pharmacovigilance case triage and signal detection notes.