Observed arrival · 2026-11-14
Lindot trains small code models for Rust
Lindot publishes open code models fine-tuned for Rust, with model weights, benchmark comparisons, and a description of its training and evaluation methods.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Rust developers testing or running open code models
- Worth noticing
- The evaluation description includes four private holdout tasks and a 12-gram overlap threshold for dropping training samples.
Field notes
The model cards pair parameter counts with two Rust benchmark scores and comparisons against each base model. The method section says data curation, decontamination, reinforcement learning, and benchmark setup are documented; the page also distinguishes four private holdout tasks from training data. A larger 9B model is still marked in training, with no results shown.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue