Hugging Face
Search models, datasets, users...
Ctrl K
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
HelloSun
/
fint-epqv2
Like
2
llama.cpp
flint
Mixture of Experts
expert-offloading
cpu-inference
License:
mit
Model card
Files
Files and versions
xet
Community
Copy to bucket
c255fee
fint-epqv2
2.51 MB
Ctrl+K
Ctrl+K
1 contributor
History:
16 commits
HelloSun
Add full report (moved from model card)
c255fee
verified
24 days ago
docs
Add full report (moved from model card)
24 days ago
results
Replace all results with the current build (94559012): removed superseded data
24 days ago
src
Hot/cold expert analysis: skew, profile pinning hit rates; fix hit counting and default pin fraction
24 days ago
tools
Fix runmem threshold unit bug; re-run all memory tiers enforced (native, v1, v2, v2 forced); add native data and forced-vs-designed explanation
24 days ago
.gitattributes
Safe
1.52 kB
initial commit
24 days ago
README.md
2.75 kB
Minimal model card: install and native/v1/v2 defaults; full report moved to docs/REPORT.md
24 days ago
bench.sh
Safe
1.23 kB
Upload v2 memory tiers (default and forced row-parallel) with path counters
24 days ago
flint-epq.patch
Safe
210 kB
Hot/cold expert analysis: skew, profile pinning hit rates; fix hit counting and default pin fraction
24 days ago