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) | 1.59x | 37.79x | 0.04x | 2390 |
| Medium (1K) | 1.22x | 38.46x | 0.11x | 2390 |
| Large (10K) | 1.37x | 2158.64x | 0.07x | 2379 |
Small (100) — by Category
| Category | Avg Speedup | Count |
|---|---|---|
| creation | 1.69x | 213 |
| arithmetic | 1.31x | 295 |
| math | 0.75x | 125 |
| trig | 0.70x | 216 |
| gradient | 4.26x | 22 |
| linalg | 1.39x | 269 |
| reductions | 2.51x | 413 |
| manipulation | 2.12x | 231 |
| io | 3.62x | 66 |
| indexing | 1.20x | 115 |
| bitwise | 1.58x | 10 |
| sorting | 0.93x | 75 |
| logic | 2.18x | 142 |
| statistics | 3.56x | 26 |
| sets | 2.89x | 33 |
| random | 1.13x | 46 |
| polynomials | 3.15x | 27 |
| fft | 1.49x | 66 |
Medium (1K) — by Category
| Category | Avg Speedup | Count |
|---|---|---|
| creation | 1.31x | 213 |
| arithmetic | 1.28x | 295 |
| math | 0.61x | 125 |
| trig | 0.55x | 216 |
| gradient | 3.54x | 22 |
| linalg | 1.91x | 269 |
| reductions | 1.04x | 413 |
| manipulation | 1.87x | 231 |
| io | 2.19x | 66 |
| indexing | 0.71x | 115 |
| bitwise | 0.83x | 10 |
| sorting | 0.87x | 75 |
| logic | 2.33x | 142 |
| statistics | 1.55x | 26 |
| sets | 2.25x | 33 |
| random | 0.89x | 46 |
| polynomials | 3.19x | 27 |
| fft | 0.84x | 66 |
Large (10K) — by Category
| Category | Avg Speedup | Count |
|---|---|---|
| creation | 3.70x | 213 |
| arithmetic | 1.84x | 295 |
| math | 0.73x | 125 |
| trig | 0.62x | 216 |
| gradient | 8.82x | 22 |
| linalg | 1.73x | 269 |
| reductions | 0.58x | 413 |
| manipulation | 2.17x | 231 |
| io | 2.36x | 55 |
| indexing | 0.55x | 115 |
| bitwise | 0.73x | 10 |
| sorting | 1.07x | 75 |
| logic | 4.46x | 142 |
| statistics | 1.06x | 26 |
| sets | 5.61x | 33 |
| random | 0.87x | 46 |
| polynomials | 3.11x | 27 |
| fft | 1.24x | 66 |