Spaces:
Sleeping
Sleeping
Deploy SAM-AI README.md
Browse files
README.md
CHANGED
|
@@ -1,13 +1,47 @@
|
|
| 1 |
---
|
| 2 |
-
title:
|
| 3 |
-
emoji:
|
| 4 |
-
colorFrom:
|
| 5 |
-
colorTo:
|
| 6 |
sdk: gradio
|
| 7 |
-
sdk_version:
|
| 8 |
-
python_version: '3.12'
|
| 9 |
app_file: app.py
|
| 10 |
pinned: false
|
|
|
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
title: SAM-AI Sovereign Assistant
|
| 3 |
+
emoji: 🧠
|
| 4 |
+
colorFrom: indigo
|
| 5 |
+
colorTo: purple
|
| 6 |
sdk: gradio
|
| 7 |
+
sdk_version: 4.44.0
|
|
|
|
| 8 |
app_file: app.py
|
| 9 |
pinned: false
|
| 10 |
+
license: apache-2.0
|
| 11 |
+
short_description: SAM-AI Sovereign Assistant on DeepSeek-R1-Distill-7B
|
| 12 |
---
|
| 13 |
|
| 14 |
+
# SAM-AI Sovereign Assistant (ZeroGPU Cloud Edition)
|
| 15 |
+
|
| 16 |
+
Welcome to the official cloud demonstration of **SAM-AI**, developed by **Parallax Intelligence Lab**.
|
| 17 |
+
|
| 18 |
+
## Architecture Overview
|
| 19 |
+
|
| 20 |
+
- **Base Model:** `deepseek-ai/DeepSeek-R1-Distill-Qwen-7B` (Sovereign Open-Weight Foundation)
|
| 21 |
+
- **Active Adapter:** None (Pure Foundation Model; Experiment 001 adapter rejected)
|
| 22 |
+
- **Runtime Compute:** Hugging Face Spaces **ZeroGPU** (NVIDIA RTX Pro 6000 Blackwell, 48 GB VRAM allocation)
|
| 23 |
+
- **Sovereignty Invariant:** Zero commercial third-party LLM APIs. All generation is executed directly by the loaded open-weight model and adapter on the assigned GPU.
|
| 24 |
+
|
| 25 |
+
## Key Capabilities
|
| 26 |
+
|
| 27 |
+
1. **SAM-AI Chat:**
|
| 28 |
+
- Multi-turn conversational interface with deep reasoning `<think> ... </think>` visibility.
|
| 29 |
+
- Mathematics, multi-step problem solving, and instruction following.
|
| 30 |
+
- Streaming token output with latency and generation status.
|
| 31 |
+
|
| 32 |
+
2. **SAM-AI Code:**
|
| 33 |
+
- Dedicated coding assistant tab using the identical verified model.
|
| 34 |
+
- Code generation, syntax analysis, refactoring recommendations, and unified diff proposals.
|
| 35 |
+
- Bounded, safe execution without unrestricted shell or file system writes.
|
| 36 |
+
|
| 37 |
+
## ZeroGPU Free Tier Guidelines
|
| 38 |
+
|
| 39 |
+
This Space operates on Hugging Face's shared free ZeroGPU infrastructure:
|
| 40 |
+
- **Allocation:** Up to 48 GB VRAM dynamically provisioned during inference calls.
|
| 41 |
+
- **Generation Duration:** Bounded to 60 seconds per response to maintain fair queue scheduling.
|
| 42 |
+
- **Fair Use:** Free Hugging Face user accounts receive 5 minutes of daily GPU time; unauthenticated users receive 2 minutes. When daily quotas are exhausted, requests enter a wait state until the next 24-hour cycle.
|
| 43 |
+
|
| 44 |
+
## Configuration & Security
|
| 45 |
+
|
| 46 |
+
- **Adapter Protection:** To protect proprietary weights, the adapter can be loaded from a private Hugging Face model repository by setting the `HF_TOKEN` secret in Space Settings, or bundled directly in the private Space repository.
|
| 47 |
+
- **Fail-Closed Verification:** On startup, all adapter tensors and configuration files are cryptographically verified against SHA-256 manifests. The Space fails closed with structured error reporting if files are missing or tampered with.
|