Decision Workspace
think-scrubber vs mdstream vs lol_html
Side-by-side comparison of Rust crates
48
think-scrubber
experimentalv0.1.0
Stateful streaming parser to scrub/extract thinking blocks in LLM completions.
54
mdstream
experimentalv0.2.0
Streaming-first Markdown middleware for LLM output (committed + pending blocks, render-agnostic).
79
lol_html
stablev2.9.0
Streaming HTML rewriter/parser with CSS selector-based API
Core Metrics
| think-scrubber | mdstream | lol_html | |
|---|---|---|---|
| Health Score | 48 | 54 | 79 |
| Total Downloads | 12 | 40.8K | 3.4M |
| 30d Downloads | 0 | 11.6K | 312.6K |
| Dependents | 0 | 1 | 4.3K |
| Releases | 1 | 2 | 25 |
| Last Updated | 5d ago | 147d ago | 19d ago |
| Age | 5d | 5m | 6y 6m |
Health Breakdown
think-scrubber
Maintenance
13
Quality
15
Community
6
Popularity
1
Documentation
13
mdstream
Maintenance
10
Quality
16
Community
7
Popularity
6
Documentation
15
lol_html
Maintenance
18
Quality
18
Community
20
Popularity
8
Documentation
15
Technical Details
| think-scrubber | mdstream | lol_html | |
|---|---|---|---|
| Version | 0.1.0 | 0.2.0 | 2.9.0 |
| Stable (≥1.0) | ✗ No | ✗ No | ✓ Yes |
| License | Apache-2.0 | MIT OR Apache-2.0 | BSD-3-Clause |
| Dependencies | 0 | 1 | 22 |
| Crate Size | 9KB | 63KB | 150KB |
| Features | 0 | 3 | 2 |
| Yanked % | 0.0% | 0.0% | 0.0% |
| Edition | 2021 | 2024 | 2024 |
| MSRV | — | 1.85 | 1.85 |
| Owners | 1 | 1 | 2 (team) |
Links
Quick Verdict
- •lol_html leads with a health score of 79/100, but none of the options score above 80.
- •lol_html is depended on by 4.3K crates — strongest ecosystem trust.
- •think-scrubber, mdstream are pre-1.0 — API may change.