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
57.8
TFLOPS
Lucas Pirola
PRO
pirola
36
4
7
Follow
dark-pen's profile picture
webbrain-one-724833's profile picture
Bbbbbnnn's profile picture
11 followers
ยท
51 following
AI & ML interests
None yet
Recent Activity
reacted
to
comgen42
's
post
with ๐ฅ
29 days ago
๐ป Kodiak-v0.3-1B: seven new kinds of decision, measured on real data. Kodiak is an open 1B encoder that answers typed questions with calibrated confidence, or says "can't tell". New in v0.3: โข pairwise judge (which answer is better?) โข long-answer hallucination checks โข stance and sarcasm โข policy violation and refund-eligibility checks against your written rules โข agent step safety (run it, ask first, or never) We stopped trusting our own synthetic tests and judged v0.3 on real labelled data it never trained on (RAGBench, MT-Bench human judgments, SemEval stance): 0.21 โ 0.36, averaged over 3 training runs. Nothing else got worse, and when it says "can't tell" it's right 93% of the time (up from 88%). Known limits are in the model card. ๐ฆ https://huggingface.co/cortex-agent-llc/kodiak-v0.3-1b ๐ฏ Accuracy mode: https://huggingface.co/cortex-agent-llc/kodiak-v0.3-1b-accuracy ๐น๏ธ Demo: https://huggingface.co/spaces/comgen42/kodiak-demo ๐ What we built, and the kill that changed our process: https://cortexagent.com/blog/kodiak-v0-3-seven-new-kinds-of-decision-measured-on-real-data
published
a model
about 1 month ago
pirola/Ornith-1.5-35B-A3B-exl3-4.0bpw-hq
published
a model
about 1 month ago
pirola/Ornith-1.5-35B-A3B-exl3-5.0bpw-hq
View all activity
Organizations
pirola
's Spaces
2
Sort:ย Recently updated
Sleeping
Agents
Trackio
๐ฏ
Sleeping
Agents
First Agent Template
โก
Find the current time in any timezone