diff --git a/hugo.toml b/hugo.toml
index 306c938..02b03d5 100644
--- a/hugo.toml
+++ b/hugo.toml
@@ -36,7 +36,7 @@ disableKinds = ['taxonomy', 'term', 'rss', 'sitemap', '404']
[[params.repos]]
name = "ndarray"
- kind = "library · v0.41.0"
+ kind = "library · v0.41.1"
href = "https://github.com/go-ndarray/ndarray"
blurb = "The library, in 13 dtypes with NumPy 2 promotion. Creation (Zeros/Arange/Linspace/…), strided views, broadcasting elementwise + ufuncs, reductions (Sum in NumPy's pairwise order, within a stated error bound checked against math/big, from go-simd/floats; Max/ArgMax/CumSum/Clip/Where), manipulation (Reshape/Concatenate/Stack), linalg (MatMul/Dot) via a panel-packed cache-blocked GEMM with a go-asmgen SIMD-FMA micro-kernel (NEON 4×8 on arm64, AVX2/FMA 6×8 on amd64, VSX 8×8 on ppc64le, LASX 8×8 on loong64, RVV 6×16, or 8×16 on 256-bit vectors, on riscv64, 4×8 on s390x), and a Workspace arena for loops (Use/Reset/Detach). Packages linalg, fft and random are numpy.linalg, numpy.fft and numpy.random."
[[params.repos]]
diff --git a/layouts/index.html b/layouts/index.html
index d32be7d..70c7d54 100644
--- a/layouts/index.html
+++ b/layouts/index.html
@@ -82,7 +82,7 @@
CGO_ENABLED=0 across the org. No cgo, no BLAS shim.
Gated in CI, differentially checked against numpy.
Row-major strided arrays, NumPy broadcasting, and NumPy basic-indexing Slice returning strided views that share data.
Workspace arena for loopsBind the inputs with ws.Use, compute as usual, ws.Reset() at the end of each pass: results stop costing the garbage collector anything. On a Zen 3, x + y on 1 024 elements drops from 2.6 to 0.38 µs (2.1× NumPy).
Workspace arena for loopsBind the inputs with ws.Use, compute as usual, ws.Reset() at the end of each pass: results stop costing the garbage collector anything. On a Zen 3, x + y on 1 024 elements drops from 2.6 to 0.38 µs, Go against Go.
go-asmgen SIMD kernels on amd64 (SSE2, AVX2/FMA picked at run time), arm64 (NEON), ppc64le (VSX), loong64 (LASX, when the CPU has it), riscv64 (RVV, when the CPU has it) and s390x (vector facility, measured on a real IBM z17); the 32-bit targets run the same pure-Go code those kernels are tested against.
Exp/LogPorts of Arm's optimized-routines: worst 0.504 / 0.508 ULP against a 300-bit reference, and correct where Go's amd64 math.Exp returns +Inf (golang/go#81995) and math.Log is wrong on subnormals (golang/go#56600).
The suite runs on amd64, arm64 and 386 natively and on riscv64/loong64/ppc64le/s390x/arm under qemu, on Linux, macOS and Windows, and compiles for every GOOS/GOARCH pair Go supports; v0.2.0 was also run on real amd64, arm64, ppc64le, riscv64 and loong64 hardware.
Sum/Mean/Max at 4 Mi (NumPy 2.5.3, 16-core Zen 3)MatMul from 512² (Zen 3)Sum 0.70×, fft.FFT 0.54× NumPy's time)Exp/Log, ports of Arm's optimized-routinesMeasured, wins and losses. Against NumPy 2.5.3 + OpenBLAS 0.3.34 on a 16-core AMD Zen 3: faster on whole-array Sum/Mean/Max at 4 Mi (2.3–4.5×), SumAxis(1) (2.2×), MaxAxis(1) (1.7×), Exp from 256 Ki elements (2.4–5.6×) and MatMul against single-threaded OpenBLAS (3.7–5.9× from 512²); slower against 16-thread OpenBLAS (MatMul 0.42× at 512² and 0.51× at 1024², though 1.44× faster at 256²; Dot 0.69× and mat·vec 0.52×, all measured in one run with NumPy) and on allocating elementwise forms outside a Workspace. On Apple silicon the packed GEMM reaches tuned-BLAS parity at 1024² and beats the pure-Go gonum 4–10×. On the other targets, against its own scalar code on one core: on a RISC-V SpacemiT X60 the RVV kernels make Sum up to 5.8× faster and cut MatMul time by 31–45%; on an IBM z17 the s390x kernels take MatMul 1024² from 340 to 70 ms and cut Max by 92–96%.
SqrtInto (0.51× NumPy's time), the complex fft.FFT (0.54×), Sum (0.70×) and Exp (0.89×). Slower on 188, chiefly from:
+Add 3.1× NumPy's time, where AddInto ties);Ceil 20×, Abs 12×);linalg at 64² (1.4–4×);random draws built on log/exp.gonum 4–10×. On the other targets, against its own scalar code on one core: on a RISC-V SpacemiT X60 the RVV kernels make Sum up to 5.8× faster and cut MatMul time by 31–45%; on an IBM z17 the s390x kernels take MatMul 1024² from 340 to 70 ms and cut Max by 92–96%.No cgo, no BLAS shim. A portable pure-Go core, multicore, with SIMD kernels on amd64, arm64, ppc64le, loong64, riscv64 and s390x and a Workspace arena that takes the garbage collector out of loops. Faster than NumPy on whole-array and row reductions, on large Exp and on MatMul against single-threaded BLAS, slower against 16-thread BLAS — the numbers are on the performance page.
No cgo, no BLAS shim. A portable pure-Go core, multicore, with SIMD kernels on amd64, arm64, ppc64le, loong64, riscv64 and s390x and a Workspace arena that takes the garbage collector out of loops. Measured against NumPy one core against one core: faster on Sum, Exp, SqrtInto and the complex FFT, slower on 188 of 256 operations, which are being worked through. The numbers are on the performance page.