All benchmarks measure computation time from JS and Python, respectively. To learn more, check out benchmark methodology.
Size Scaling Summary
| Array Size | Avg Speedup | Best Case | Worst Case | Benchmarks |
|---|---|---|---|---|
| Small (100) | 1.12x | 40.38x | 0.09x | 2390 |
| Medium (1K) | 1.02x | 77.99x | 0.12x | 2390 |
| Large (10K) | 1.20x | 2422.82x | 0.10x | 2379 |
Small (100) — by Category
| Category | Avg Speedup | Count |
|---|---|---|
| creation | 1.35x | 213 |
| arithmetic | 0.72x | 295 |
| math | 0.61x | 125 |
| trig | 0.59x | 216 |
| gradient | 3.70x | 22 |
| linalg | 1.45x | 269 |
| reductions | 1.72x | 413 |
| manipulation | 1.06x | 231 |
| io | 3.04x | 66 |
| indexing | 0.89x | 115 |
| bitwise | 0.55x | 10 |
| sorting | 0.67x | 75 |
| logic | 0.94x | 142 |
| statistics | 3.03x | 26 |
| sets | 2.78x | 33 |
| random | 1.03x | 46 |
| polynomials | 2.13x | 27 |
| fft | 1.16x | 66 |
Medium (1K) — by Category
| Category | Avg Speedup | Count |
|---|---|---|
| creation | 1.22x | 213 |
| arithmetic | 0.99x | 295 |
| math | 0.58x | 125 |
| trig | 0.57x | 216 |
| gradient | 3.42x | 22 |
| linalg | 1.69x | 269 |
| reductions | 0.91x | 413 |
| manipulation | 1.07x | 231 |
| io | 2.32x | 66 |
| indexing | 0.62x | 115 |
| bitwise | 0.49x | 10 |
| sorting | 0.89x | 75 |
| logic | 1.39x | 142 |
| statistics | 1.46x | 26 |
| sets | 2.46x | 33 |
| random | 0.89x | 46 |
| polynomials | 2.14x | 27 |
| fft | 0.70x | 66 |
Large (10K) — by Category
| Category | Avg Speedup | Count |
|---|---|---|
| creation | 3.29x | 213 |
| arithmetic | 1.61x | 295 |
| math | 0.66x | 125 |
| trig | 0.66x | 216 |
| gradient | 7.11x | 22 |
| linalg | 1.37x | 269 |
| reductions | 0.57x | 413 |
| manipulation | 1.37x | 231 |
| io | 2.84x | 55 |
| indexing | 0.52x | 115 |
| bitwise | 1.01x | 10 |
| sorting | 0.93x | 75 |
| logic | 3.67x | 142 |
| statistics | 0.82x | 26 |
| sets | 5.39x | 33 |
| random | 0.81x | 46 |
| polynomials | 1.93x | 27 |
| fft | 0.94x | 66 |