carl-studio
CARL (Coherence-Aware Reinforcement Learning) — information-theoretic reward signals for LLM training via token-level probability distributions
- Latest release
- May 09, 2026
- Releases
- 6
- Known CVEs
- 0
- First release
- Apr 09, 2026
- License
- MIT
Repository
Source
- Stars
- —
- Forks
- —
- Open issues
- —
Security score
No OpenSSF Scorecard available for this repository.
Packages from this repo
Insights
Activity
- Total releases
- 6
- Last 12 months
- 6
- Cadence
- ~daily
- Dependencies
- 41
Releases per month
last 12 monthsRelease mix
- minor 2
- patch 3
6
releases
Dependencies
Depends on
0.20.1-
a2a-python
>=0.0.1
-
anthropic
>=0.95
-
anyio
>=4.0
-
arxiv
>=3.0
-
bitsandbytes
-
carl-core
>=0.2.0
-
coinbase-agentkit
>=0.7
-
cryptography
>=42.0.0
-
datasets
>=4.0
-
gradio-client
>=1.0
1–10 of 41
Used by
Nothing tracked depends on this yet.
Releases
| Version | Released | |
|---|---|---|
0.20.1
patch
| ||
0.20.0
minor
| ||
0.18.3
patch
| ||
0.18.2
patch
| ||
0.18.1
minor
| ||
0.3.0
initial
|