Latest generation only (zen5-cpu, nvidia-gpu). Each entry is a commit on the main branch that
added or changed a report, newest first, with the reports present at that commit. Links are
permanent: they point at the commit. Older hardware generations stay in the repository and its git
history but are not listed here. Back to the current results.
9f0ea6ee38Publish zen5-cpu results at tenferro-rs 8fec1ca (#276) tenferro-rs 8fec1ca70852
New or changed (9)
7930c010f3Publish verified A100 solve/QR readback latency results (#272) tenferro-rs 234a51369c25
New or changed (1)
50a7dc9b3eAccept benchmark run 40077067870 attempt 1 tenferro-rs 06ac2475a286
New or changed (4)
12e0b50f98Accept benchmark run 39735132118 attempt 1 tenferro-rs b59cf27cbfea
New or changed (4)
b3f752b55eMerge pull request #226 from tensor4all/feat/gpu-throughput-mode tenferro-rs e921b059a5e3
New or changed (2)
ab5db732ecAccept benchmark run 39039069086 attempt 1 tenferro-rs 4de2e1226337
New or changed (4)
7a244ea85fMerge pull request #202 from tensor4all/bench/gpu-dashboard tenferro-rs 88daa885a5bb
New or changed (14)
fdf7f60bf9Merge pull request #198 from tensor4all/bench/gpu-refresh-2026-10-17
New or changed (15)
273a4770acMerge pull request #194 from tensor4all/bench/gpu-compare-refresh
New or changed (2)
883166a792Merge pull request #193 from tensor4all/bench/gpu-vendor-sampling
New or changed (8)
0d98678c1dAdd balanced CUDA A/A and declared paired confirmation (#192)
New or changed (2)
f9ebff8eacInclude public CUDA latency and state controls in routine runs (#191)
New or changed (7)
2ff7c9ae88Separate routine CUDA coverage from sampling effort (#189)
New or changed (5)
a889659e86Measure CUDA command and phase wall costs before routine-profile changes (#187)
New or changed (1)
656d77453cPublish CUDA shape-stream scan and separate public counter observations (#186)
New or changed (1)
d20ae8049fMerge pull request #182 from tensor4all/bench/gpu-reshape-stack
New or changed (2)
61e07ea284Merge pull request #181 from tensor4all/bench/gpu-integer-operations
New or changed (1)
2035a8fb77Merge pull request #180 from tensor4all/bench/gpu-read-operations
New or changed (2)
8e9aba4e04Merge pull request #179 from tensor4all/bench/gpu-scalar-scale
New or changed (2)
33817711beMerge pull request #178 from tensor4all/bench/gpu-affine-normalization
New or changed (1)
43e9bde3cfMerge pull request #176 from tensor4all/bench/gpu-pad-domains
New or changed (1)
6edf691537Merge pull request #175 from tensor4all/bench/gpu-lifetime-counters
New or changed (1)
81d4e665b5Merge pull request #174 from tensor4all/bench/gpu-compare-lifetime
New or changed (1)
4056dde9e6Merge pull request #173 from tensor4all/bench/gpu-prepared-einsum
New or changed (1)
fbe007a409Merge pull request #172 from tensor4all/bench/gpu-general-dot
New or changed (1)
7febb22bfbMerge pull request #171 from tensor4all/bench/gpu-axpby-coefficients
New or changed (1)
4e7e520cbfMerge pull request #170 from tensor4all/bench/gpu-lstsq-contract
New or changed (1)
ed8e415034Merge pull request #169 from tensor4all/bench/gpu-lu-outputs
New or changed (1)
d620e8a3ccMerge pull request #168 from tensor4all/bench/gpu-strided-read
New or changed (1)
d2bcd32cc0Merge pull request #167 from tensor4all/bench/gpu-dot-accumulation
New or changed (1)
258359dbf0Merge pull request #166 from tensor4all/bench/gpu-accumulation
New or changed (1)
63eeec296dMerge pull request #164 from tensor4all/bench/gpu-householder-state
New or changed (1)
fb327fb83cVerify all GPU outputs for full SVD, slogdet and compact QR (#163)
New or changed (1)
5fb5897b59Add GPU reduction, materialization and indexed-operation comparisons (#162)
New or changed (2)
4849e1fee6Verify GPU linalg AD and add matched JAX JIT comparisons (#161)
New or changed (2)
19d3093d1cExtend GPU public API coverage and verified JAX comparisons (#160)
New or changed (10)
840f41726aBenchmark six CUDA public API families with PyTorch and JAX (#158)
New or changed (6)
6c62be9863Merge pull request #132 from tensor4all/gpu/perf-issues-latest
New or changed (1)
f5f7714ffaMerge pull request #116 from tensor4all/t4a-2010-final-fixes
New or changed (9)
dd129e7755Merge pull request #102 from tensor4all/agent/refresh-nvidia-gpu-main-20260918
New or changed (8)
8d1e3b4dd9Merge pull request #99 from tensor4all/agent/refresh-nvidia-gpu-main-20260915
New or changed (6)
ce729ecfa0Merge pull request #93 from tensor4all/agent/update-nvidia-gpu-results-20260822
New or changed (5)
da4a7a9722Merge pull request #69 from tensor4all/agent/scale-gpu-permutation-benchmarks
New or changed (1)
30d12bd611Merge pull request #68 from tensor4all/agent/update-nvidia-gpu-benchmarks
New or changed (6)
de24290b46Merge pull request #57 from tensor4all/result-refresh-tenferro-3abc910
New or changed (6)
a296e443bcMerge pull request #54 from tensor4all/full-result-refresh
New or changed (6)
015883bcb1Merge pull request #47 from tensor4all/gpu-permutation-suite
New or changed (1)
8bde6d5e20Merge pull request #40 from tensor4all/codex/add-gpu-linalg-jvp-vjp-benchmark
New or changed (1)
4b791d16beMerge pull request #38 from tensor4all/codex/update-nvidia-gpu-benchmarks-d5c768c7
New or changed (4)
f45156fe0bMerge pull request #39 from tensor4all/codex/add-gpu-linalg-jvp-vjp-benchmark
New or changed (1)
b9af9c8e68Merge pull request #24 from tensor4all/gpu/bench/rtx-3060
New or changed (4)
5ca74439a2Update GPU benchmark reports with new device information and results
New or changed (4)
36db3b443cUpdate development environment configuration and benchmark results
New or changed (1)
4c5a906450Record GPU device metadata in GPU benchmark reports.
New or changed (1)
4cd7ff3b08Run gpu/tensornetwork serially and disable JAX temporarily.
New or changed (1)
278b483174Merge pull request #23 from tensor4all/feat/gpu-tensornetwork-benchmarks
New or changed (1)
38236ee90dAdd larger GPU solve benchmarks
New or changed (3)