llama.cppvsretroassembly
Two self-hosted projects, compared on the numbers selfhost.directory tracks live: score, popularity, release cadence and setup effort.
| llama.cpp | retroassembly | |
|---|---|---|
| Score | 93/100 | 56/100 |
| Rank in category | #2 of 25 in LLM Inference | #5 of 5 in Emulation |
| GitHub stars | 132 714 | 1 153 |
| Stars, last 30 days | +3 259 | +72 |
| Stable releases, 12 months | 1467 | 3 |
| Latest release | b11981 · 29 Oct 2026 | v7.260719.1051 · 19 Jul 2026 |
| Setup | Moderate | Easy |
| Official Docker image | Yes | Yes |
| License | MIT | MIT |
| Main language | C++ | Typescript |
Which one should you pick?
llama.cpp scores 93/100 against 56/100 for retroassembly, so it is the safer default.
llama.cpp ships more often: 1467 stable releases in the past year, against 3 for retroassembly.
If you want the quickest setup, retroassembly is easier to run; llama.cpp needs more configuration.
llama.cpp is growing faster right now (+3 259 stars in 30 days).
llama.cpp details →retroassembly details →