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) | 1.17x | 37.83x | 0.10x | 2390 |
| Medium (1K) | 1.00x | 39.58x | 0.11x | 2390 |
| Large (10K) | 1.23x | 2336.20x | 0.11x | 2379 |
Small (100) — by Category
| Category | Avg Speedup | Count |
|---|---|---|
| creation | 1.48x | 213 |
| arithmetic | 0.78x | 295 |
| math | 0.61x | 125 |
| trig | 0.59x | 216 |
| gradient | 3.85x | 22 |
| linalg | 1.46x | 269 |
| reductions | 1.91x | 413 |
| manipulation | 1.13x | 231 |
| io | 3.39x | 66 |
| indexing | 0.89x | 115 |
| bitwise | 0.54x | 10 |
| sorting | 0.65x | 75 |
| logic | 0.99x | 142 |
| statistics | 2.93x | 26 |
| sets | 2.43x | 33 |
| random | 1.00x | 46 |
| polynomials | 2.04x | 27 |
| fft | 1.11x | 66 |
Medium (1K) — by Category
| Category | Avg Speedup | Count |
|---|---|---|
| creation | 1.19x | 213 |
| arithmetic | 1.00x | 295 |
| math | 0.56x | 125 |
| trig | 0.55x | 216 |
| gradient | 3.34x | 22 |
| linalg | 1.53x | 269 |
| reductions | 0.91x | 413 |
| manipulation | 1.07x | 231 |
| io | 2.17x | 66 |
| indexing | 0.59x | 115 |
| bitwise | 0.47x | 10 |
| sorting | 0.80x | 75 |
| logic | 1.49x | 142 |
| statistics | 1.34x | 26 |
| sets | 2.05x | 33 |
| random | 0.85x | 46 |
| polynomials | 2.00x | 27 |
| fft | 0.67x | 66 |
Large (10K) — by Category
| Category | Avg Speedup | Count |
|---|---|---|
| creation | 3.25x | 213 |
| arithmetic | 1.67x | 295 |
| math | 0.64x | 125 |
| trig | 0.65x | 216 |
| gradient | 5.78x | 22 |
| linalg | 1.37x | 269 |
| reductions | 0.58x | 413 |
| manipulation | 1.42x | 231 |
| io | 2.77x | 55 |
| indexing | 0.53x | 115 |
| bitwise | 0.98x | 10 |
| sorting | 0.99x | 75 |
| logic | 4.45x | 142 |
| statistics | 0.99x | 26 |
| sets | 5.49x | 33 |
| random | 0.83x | 46 |
| polynomials | 1.98x | 27 |
| fft | 1.03x | 66 |