File size: 3,201 Bytes
0d37852 340d68e 0d37852 49735ac 0d37852 6697a55 0d37852 340d68e 0d37852 340d68e 0d37852 6697a55 49735ac 0d37852 49735ac 6697a55 49735ac 0d37852 340d68e 0d37852 340d68e 0d37852 49735ac 0d37852 49735ac 0d37852 49735ac 340d68e 49735ac 0d37852 49735ac 0d37852 49735ac 0d37852 49735ac 0d37852 49735ac 0d37852 49735ac 340d68e 0d37852 49735ac 0d37852 49735ac 0d37852 49735ac 0d37852 49735ac 75f012d 340d68e 2197abe 340d68e 2197abe 340d68e | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 | ---
license: mit
language:
- en
tags:
- safety
- crisis-detection
- text-classification
- mental-health
- gguf
- llama.cpp
- ollama
base_model: Qwen/Qwen3-4B
pipeline_tag: text-generation
---
# NOPE Edge GGUF (4B)
GGUF quantized version of [nopenet/nope-edge](https://huggingface.co/nopenet/nope-edge) for local inference with Ollama and llama.cpp.
> **License:** [MIT](LICENSE.md) - free for any use, including commercial. Built on Qwen3 (Apache-2.0); see NOTICE.md.
---
## Quick Start with Ollama
```bash
# Download the GGUF and Modelfile
huggingface-cli download nopenet/nope-edge-GGUF nope-edge-q8_0.gguf Modelfile --local-dir .
# Create Ollama model
ollama create nope-edge -f Modelfile
# Run inference
ollama run nope-edge "I can't take this anymore"
```
---
## Files
| File | Quantization | Size |
|------|--------------|------|
| `nope-edge-q8_0.gguf` | Q8_0 | 3.99 GB |
---
## Output Format
The model outputs XML with chain-of-thought reasoning:
**Crisis detected:**
```xml
<reflection>User expresses direct suicidal intent with timeline...</reflection>
<risks>
<risk subject="self" type="suicide" severity="high" imminence="urgent" features="active_ideation,timeline_stated"/>
</risks>
```
**No crisis:**
```xml
<reflection>Gaming slang, no genuine crisis indicators...</reflection>
<risks/>
```
---
## Risk Types
| Type | Description |
|------|-------------|
| `suicide` | Suicidal ideation, plans, or intent |
| `self_harm` | Non-suicidal self-injury |
| `self_neglect` | Eating disorders, medical neglect |
| `violence` | Threats toward others |
| `abuse` | Domestic/intimate partner violence |
| `sexual_violence` | Sexual assault, coercion |
| `exploitation` | Trafficking, grooming, sextortion |
| `stalking` | Persistent unwanted contact |
| `neglect` | Child or elder neglect |
---
## Hardware Requirements
| Model | Quant | RAM/VRAM | CPU Latency | GPU Latency |
|-------|-------|----------|-------------|-------------|
| nope-edge (4B) | Q8_0 | ~5GB | ~2s | ~200ms |
| nope-edge-mini (1.7B) | Q8_0 | ~2GB | ~1s | ~100ms |
---
## Source Model
- **Repository**: [nopenet/nope-edge](https://huggingface.co/nopenet/nope-edge)
- **Base**: Qwen/Qwen3-4B
- **Purpose**: Mental health crisis classification
---
## Important
- **Not a medical device.** Outputs are probabilistic signals for triage, not clinical assessments.
- **False positives and negatives will occur.** Use for flagging, not autonomous decisions.
- **`features` are heuristic** — lower-confidence than type/severity and may include labels not supported by the text; treat as hints only.
- **Classification only.** Crisis resources require the [NOPE Evaluate API](https://nope.net/evaluate).
**Edge is a detection aid — not a predictive, diagnostic, or therapeutic tool, and not a replacement for clinical judgment.** False positives and false negatives will occur — some people in genuine crisis will not be identified — so never use Edge as the sole basis for an intervention decision, and always keep a human in the loop. If anyone is in immediate danger, contact your local emergency services or find resources at talk.help.
Full disclaimer: see DISCLAIMER.md.
|