magnitudedev / magnitude
Author description: Open source inference engine that tunes its kernels on your hardware to find the fastest way to run each model. 2x faster than Ollama. Works on Apple Silicon, NVIDIA, AMD, Intel, or just a CPU.
- Repository stars
- 7,719All-time #7,354
- Fork
- 555
- Last push
- updated today
- Repository created
- since 2026
About
Summarised by AI from the README; the original prevails
Author description: Author description: Open source inference engine that tunes its kernels on your hardware to find the fastest way to run each model. 2x faster than Ollama. Works on Apple Silicon, NVIDIA, AMD, Intel, or just a CPU.
Basic listing based on official public information; functionality has not been independently evaluated.
- Platform
- Not verified
- Pricing
- Pricing not verifiedCheck current official software, hosting, model and infrastructure pricing.
- Maintenance notes
- Not verified