diff --git a/docs/plans/counter-asic-2-status.md b/docs/plans/counter-asic-2-status.md
index f08f00c63..dfc6fb525 100644
--- a/docs/plans/counter-asic-2-status.md
+++ b/docs/plans/counter-asic-2-status.md
@@ -523,3 +523,5 @@ build-20261005-221237 (main d233fa1, fork 89dfcb95, 185 s): every stage ok; kasp
 docs/spec/01-lottery-hash.md on ca2-coord: 1.8.5 (the mixer x8 form with the measured costs), 1.13.1 (the era draw: stride, interleave, windows, the devnet stand-in, the measured spread), 1.17 (the class v3 vectors), 1.5 (the cache note); earlier tonight 1.12 and 1.13.1 (epoch_len), 1.13.2 (R1 and the emulation rule), 1.13.3 (option C and the step mapping), 1.4.5 and 1.4.6 (generator 3, the class and era in the pack), and 4.3. Public copy: level 1 on the hero and the abstract ("Built for graphics cards. A custom chip gains under 2x, and the model and the bounty are public."), the limits bullet rewritten on the chip row (0.92x with the allowance, approximate; the margin on the numbers page; the next lever named), the level 3 table filled from the final numbers in counter-asic-2-public.md (the bench page section is written from it at the ship). Waiting: the era agent's PC 1 job (G1 on the final class and G2), the node agent's gate run 3; then the integration merge and the ship.
 
 22:19. docs/analysis/proving-methods.md (branch proving-methods e7e0db7, not consensus): top recommendation re-size SP1's own GPU server (the floor is its code: the under-20 GB panic at sp1-gpu builder.rs:37, trace buffers at the maximum shard, a CUDA mempool that never releases), S_p as the dial, one server per card on rigs; the pinned ids stay; fallback and the Apple route: RISC Zero as proof-system version 2 behind the ProofSystem seam (8 GB at po2 19, 16 GB at po2 20, shipped Metal; 3 to 4 agent days); no 12 GB card has run a prover here, so a 4070 or 3060 in the loop is the first action. That branch merges into the 0.3.11 main tree as documentation (no code). evidence.md rows 17, 18 and 19 are rewritten on the measured class v3 (ba8379d).
+
+22:20. The numbers page's Counter ASIC section is written as the bench-log entry "Counter ASIC 2.0, the numbers" (the page is built from the bench log; the litepaper links /bench#counter-asic-2-0-the-numbers); the litepaper's verifier figures moved to the v3 class (2.1 ms per warp on a loaded core, 4.8x inside the gate; the cache 512 MB from year 4). Everything public now carries the final class except the PC rows of the final-class packs and the G2 counts, which the running PC 1 job supplies.
diff --git a/site/litepaper.html b/site/litepaper.html
index 73aaa2049..3c6bce685 100644
--- a/site/litepaper.html
+++ b/site/litepaper.html
@@ -408,7 +408,7 @@ body.all .pager{display:none}
       <section id="mining">
         <h2>Mining: a program that never holds still</h2>
         <p>Every GPU chain that promised ASIC resistance shipped a fixed algorithm, and a fixed algorithm gets a chip the moment the prize pays for one. Igneum does not have a fixed algorithm.</p>
-        <p>Each hour the chain derives a seed from a locked checkpoint one epoch back, passes it through a ten-minute verifiable delay so no miner can see which program a seed implies before choosing whether to publish a block, and feeds it to a deterministic generator. The generator emits a random integer program built from what graphics cards are uniquely good at: wide parallel integer maths, shuffles between the 32 lanes of a warp, and random reads over a multi-gigabyte dataset that changes daily, so the program waits on memory latency, not on maths or bandwidth. The memory footprint and instruction count are fixed and only the maths sequence is random, so no hour favours one vendor's cards and nobody gains by grinding the seed. Miners compile the program once per hour. Anyone running a node, a wallet or an exchange checks a hash on an ordinary CPU in under ten milliseconds by simulating one warp, so nobody needs a GPU except to mine. Measured: 0.41 to 0.58 ms per warp on one Apple M5 Max core with the 256 MB cache, about 17x inside the 10 ms gate; a 2019-class laptop core is not yet measured.</p>
+        <p>Each hour the chain derives a seed from a locked checkpoint one epoch back, passes it through a ten-minute verifiable delay so no miner can see which program a seed implies before choosing whether to publish a block, and feeds it to a deterministic generator. The generator emits a random integer program built from what graphics cards are uniquely good at: wide parallel integer maths, shuffles between the 32 lanes of a warp, and random reads over a multi-gigabyte dataset that changes daily, so the program waits on memory latency, not on maths or bandwidth. The memory footprint and instruction count are fixed and only the maths sequence is random, so no hour favours one vendor's cards and nobody gains by grinding the seed. Miners compile the program once per hour. Anyone running a node, a wallet or an exchange checks a hash on an ordinary CPU in under ten milliseconds by simulating one warp, so nobody needs a GPU except to mine. Measured: 0.61 ms per warp on one Apple M5 Max core for class v2 and 2.1 ms for class v3 (the mixer at x8, 5 October 2026, a loaded core; about 1.3 ms quiet, approximate), 4.8x inside the 10 ms gate; a 2019-class laptop core is not yet measured.</p>
         <p>The hash is a lottery, not a general-purpose cryptographic hash. It has to be unpredictable per nonce, free of any shortcut cheaper than honest evaluation, and free of bias a miner can exploit. It does not need preimage or collision resistance. Open: no analysis of the lottery properties exists yet. It is the first job of the external review in phase 1, and until then the hash is a design claim backed by the measurements below.</p>
         <div class="tbl"><table>
           <thead><tr><th>Clock</th><th>What changes</th><th>Miner update needed?</th></tr></thead>
@@ -434,7 +434,7 @@ body.all .pager{display:none}
             <tr><td>Hardware it is built for</td><td>CPUs. GPUs run it badly on purpose</td><td>GPUs. Any card, any vendor. Bit-exact on Apple, NVIDIA and AMD, measured</td></tr>
             <tr><td>Random program</td><td>Per hash, interpreted in a virtual machine</td><td>Per hour, compiled to native GPU code. Per hash, the 128 dataset addresses change with the nonce</td></tr>
             <tr><td>Dataset</td><td>About 2 GB, the same size since 2019, approximate</td><td>2 GB at genesis, doubling at years 4, 12 and 28 under the step schedule; a 4 GB card mines about four years, an 8 GB card past a decade</td></tr>
-            <tr><td>Light verification</td><td>256 MB cache on a CPU, milliseconds</td><td>256 MB cache on a CPU, one warp under 10 ms, the gate. Measured 0.41 to 0.58 ms on one Apple M5 Max core; a 2019-class core not yet</td></tr>
+            <tr><td>Light verification</td><td>256 MB cache on a CPU, milliseconds</td><td>256 MB cache on a CPU (512 MB from year 4), one warp under 10 ms, the gate. Measured 2.1 ms on one loaded Apple M5 Max core for class v3; a 2019-class core not yet</td></tr>
             <tr><td>Changes over time</td><td>None. A fixed design, unchanged for seven years</td><td>A new program every hour, its memory pattern with it; era draws and reserved families on a schedule fixed at genesis. Nobody touches it</td></tr>
             <tr><td>Seed grinding</td><td>Not applicable, the program comes from the hash input</td><td>Closed by a verifiable delay between seed and program</td></tr>
             <tr><td>Useful work</td><td>None. Hashing only</td><td>NVIDIA cards with 24 GB or more prove every block and sell proofs to other chains; AMD and Apple cards mine, and a prover for them lands when a zkVM ships one</td></tr>