v-vovk commited on
Commit
210d7b1
·
verified ·
1 Parent(s): 1f9ce70

Update anonymous model card metadata

Browse files
Files changed (1) hide show
  1. README.md +36 -0
README.md ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Portable Nutrition Guidance SLM (Q4)
2
+
3
+ ## Overview
4
+
5
+ This repository hosts a quantized on-device language model artifact for nutrition-guidance chat workflows.
6
+ The artifact is packaged for local iOS runtime loading and does not require backend inference.
7
+
8
+ ## Intended use
9
+
10
+ - Personal nutrition guidance conversations
11
+ - Meal pattern coaching support
12
+ - Educational wellness tips for non-emergency scenarios
13
+
14
+ ## Safety and medical disclaimer
15
+
16
+ This model is not a diagnostic or treatment system.
17
+ It should not be used for emergency decisions, medication dosing, or definitive medical advice.
18
+ Users should consult qualified clinicians for diagnosis and treatment decisions.
19
+
20
+ ## Known limitations
21
+
22
+ - Outputs can still be incomplete, ambiguous, or incorrect.
23
+ - Responses may underperform on rare edge cases outside training distribution.
24
+ - Prompt phrasing can materially affect result quality and safety.
25
+
26
+ ## Artifact details
27
+
28
+ - Version: `medgemma-ios-q4-v9-fixture-calib`
29
+ - Download URL: `https://huggingface.co/v-vovk/medgemma-ios-q4/resolve/main/medgemma-ios-q4-v9-fixture-calib.zip`
30
+ - SHA256: `dd794a634204557bec6639351d4e3fcfaa5717aca8599f5d7af7a3b29398684b`
31
+ - Size (bytes): `2594347078`
32
+
33
+ ## Evaluation expectations
34
+
35
+ - Validate behavior against curated safety and hallucination checks before release.
36
+ - Track real-world mismatch reports and feed corrected examples into the next training cycle.