Hugging Face
Models
Datasets
Spaces
Buckets
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
nandukmelath
/
nirnay
Like
0
Zero-Shot Classification
37 datasets
English
nirnay
decision-model
text-classification
intent-classification
llm-routing
guardrails
content-moderation
spam-detection
calibrated-probabilities
few-shot
deterministic
cpu
laya-compatible
jev-compatible
Eval Results (legacy)
License:
apache-2.0
Model card
Files
Files and versions
xet
Community
Copy to bucket
main
nirnay
/
code
/
train
109 kB
Ctrl+K
Ctrl+K
1 contributor
History:
2 commits
nandukmelath
Nirnay v3.1 is the new default: request routing 0.40 -> 0.88, SMS spam, emotion, toxicity, RAG up; previous default moved to v2/
61ffaac
verified
2 days ago
bakeoff_encoders.py
Safe
7.82 kB
Nirnay v2 (bge-base, default) + v1 (fast/): weights, package, code, benchmarks
5 days ago
build_data.py
57.4 kB
Nirnay v3.1 is the new default: request routing 0.40 -> 0.88, SMS spam, emotion, toxicity, RAG up; previous default moved to v2/
2 days ago
precompute.py
Safe
9.62 kB
Nirnay v2 (bge-base, default) + v1 (fast/): weights, package, code, benchmarks
5 days ago
train_gpu.py
17 kB
Nirnay v3.1 is the new default: request routing 0.40 -> 0.88, SMS spam, emotion, toxicity, RAG up; previous default moved to v2/
2 days ago
train_head.py
Safe
17.6 kB
Nirnay v2 (bge-base, default) + v1 (fast/): weights, package, code, benchmarks
5 days ago
