igneum-labs cd5d2ca6e6 class v5: the RTX 4090 rows (the fleet's pod cv5-2, 10:23 to 10:30 UK): class v5 against class v4 on the two pinned packs, 63.058 against 63.066 MH/s (equal within 0.01 percent), the build +0.24 ms with the devnet's 93 leaves, watts a warm-up drift with the classes interleaved inside it, every dataset sample and vector warp bit-exact on the card; the hot-set bound's number and its generator-draw form recorded; the time line ... Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>		2026-10-07 09:31:22 +00:00
..
20261007T083716Z	class v5: the verifier measured on igneum-build-1 (tools/class-v5/verify-bench-remote.sh, the ladder's method, box at load 70): class v5 adds 0.28 ms cold and 0.20 ms on average over class v4 on the loaded reference core (8.74 to 9.03 ms cold with the sibling loaded), both under the 10 ms gate; the raw lines under docs/design/class-v5-bench; the page's time line with the fork suites and the packs test	2026-10-07 08:38:41 +00:00
pod-4090-20261007	class v5: the RTX 4090 rows (the fleet's pod cv5-2, 10:23 to 10:30 UK): class v5 against class v4 on the two pinned packs, 63.058 against 63.066 MH/s (equal within 0.01 percent), the build +0.24 ms with the devnet's 93 leaves, watts a warm-up drift with the classes interleaved inside it, every dataset sample and vector warp bit-exact on the card; the hot-set bound's number and its generator-draw form recorded; the time line	2026-10-07 09:31:22 +00:00