Decision Workspace
csm-core vs csm-ws vs adk-audio
Side-by-side comparison of Rust crates
41
csm-core
experimentalv0.1.0
Candle-based inference for Sesame CSM-1B text-to-speech (model + Mimi audio decoder). Derived from cartesia-one/csm.rs.
39
csm-ws
experimentalv0.1.0
Real-time WebSocket TTS server for Sesame CSM-1B, mirroring the ElevenLabs stream-input API. Built on axum and Candle.
63
adk-audio
growingv2.3.0
Audio intelligence and pipeline orchestration for ADK-Rust agents
Core Metrics
| csm-core | csm-ws | adk-audio | |
|---|---|---|---|
| Health Score | 41 | 39 | 63 |
| Total Downloads | 72 | 22 | 2.2K |
| 30d Downloads | 0 | 0 | 0 |
| Dependents | 1 | 0 | 17 |
| Releases | 1 | 1 | 16 |
| Last Updated | 117d ago | 117d ago | today |
| Age | 3m | 3m | 6m |
Health Breakdown
csm-core
Maintenance
7
Quality
15
Community
7
Popularity
2
Documentation
10
csm-ws
Maintenance
7
Quality
14
Community
6
Popularity
2
Documentation
10
adk-audio
Maintenance
19
Quality
16
Community
9
Popularity
4
Documentation
15
Technical Details
| csm-core | csm-ws | adk-audio | |
|---|---|---|---|
| Version | 0.1.0 | 0.1.0 | 2.3.0 |
| Stable (≥1.0) | ✗ No | ✗ No | ✓ Yes |
| License | AGPL-3.0-only | AGPL-3.0-only | Apache-2.0 |
| Dependencies | 14 | 18 | 29 |
| Crate Size | 44KB | 46KB | 131KB |
| Features | 6 | 6 | 19 |
| Yanked % | 0.0% | 0.0% | 0.0% |
| Edition | 2021 | 2021 | 2024 |
| MSRV | — | — | 1.95 |
| Owners | 1 | 1 | 1 |
Links
Quick Verdict
- •adk-audio leads with a health score of 63/100, but none of the options score above 80.
- •csm-core, csm-ws are pre-1.0 — API may change.