merge dev → main (exec inline + platforms null fix)

benchmarks/blast-rpc.yml

@@ -30,7 +30,7 @@ methodology:
 
 findings:
   - "{{best_name}} currently leads free Blast RPC at {{best_p50}} (`eth_getBlockByNumber` p50, 24h) across 4 measured providers."
-  - "`rpc.blast.io` posts ~2 ms in every region simultaneously. On-chain response headers confirm geo-distributed nodes: `blast_blast-mainnet_fra` from eu-west, `blast_blast-mainnet_sin` from Singapore, a US node from us-east. Blast runs RPC nodes inside the same GCP datacenters as our Railway probes, producing sub-2 ms figures that reflect co-located infrastructure rather than typical user latency. A user in Paris reaches the Frankfurt node in ~40 ms; a user outside GCP will see 30-100 ms depending on location. Heads are fresh per stale detection."
+  - "`rpc.blast.io` posts ~2 ms in every region. Response headers confirm geo-distributed GCP nodes co-located with our Railway probes (`blast_blast-mainnet_fra` from eu-west, `blast_blast-mainnet_sin` from Singapore, US node from us-east). Sub-2 ms reflects co-location, not real-world latency; outside GCP expect 30-100 ms. Heads are fresh per stale detection."
   - "{{name:drpc}} ({{p50:drpc}}) is the honest-infrastructure comparison point: anycast consistency across the three regions with real node round trips behind it, the profile that wins it 10 of the 12 expansion chains."
   - "{{name:tenderly}} sits at the expansion's familiar flat ~330 ms in all regions on Blast, single-origin routing on a gateway that is genuinely quick on the major chains."
 
@@ -44,7 +44,7 @@ faq:
   - q: "How is Blast RPC latency measured here?"
     a: "One identical JSON-RPC POST (`eth_getBlockByNumber`) every 60 seconds against each provider from each of 3 regions, with the same plain HTTP client. Wall-clock round-trip is recorded at millisecond precision; p50/p90/p99 are computed via Prometheus `quantile_over_time` over 24 hours. Responses are classified (`ok` / `http_err` / `jsonrpc_err` / `stale` / `timeout`) so an endpoint stuck on an old head or returning errors behind HTTP 200 is never ranked as fastest. The harness is open source and every number on this page is a public Prometheus query you can run yourself."
   - q: "Is rpc.blast.io really that fast, or is something else going on?"
-    a: "Blast runs geo-distributed RPC nodes inside major cloud datacenters (GCP us-east, GCP europe-west4/Frankfurt, GCP asia-southeast1/Singapore). Our Railway probes run in the same three GCP regions, so each probe hits the co-located Blast node with a sub-LAN round trip. The `x-node-id` response header confirms this: `blast_blast-mainnet_fra` from eu-west, `blast_blast-mainnet_sin` from Singapore. The blocks are fresh — this is not a cache hit — but the 1-2 ms figure reflects network co-location, not what a developer outside GCP infra will experience. From a typical residential connection or a non-GCP cloud, expect 30-100 ms."
+    a: "Blast runs geo-distributed RPC nodes inside major cloud datacenters (GCP us-east, GCP europe-west4/Frankfurt, GCP asia-southeast1/Singapore). Our Railway probes run in the same three GCP regions, so each probe hits the co-located Blast node with a sub-LAN round trip. The `x-node-id` response header confirms this: `blast_blast-mainnet_fra` from eu-west, `blast_blast-mainnet_sin` from Singapore. The blocks are fresh (not a cache hit) but the 1-2 ms figure reflects network co-location, not what a developer outside GCP infra will experience. From a typical residential connection or a non-GCP cloud, expect 30-100 ms."
 
 source: https://github.com/ChainBench/OpenChainBench/tree/main/harnesses/rpc-capabilities
 

benchmarks/monad-rpc.yml

@@ -47,7 +47,7 @@ methodology:
 
 findings:
   - "{{best_name}} currently leads the free Monad RPC field at {{best_p50}} (p50, 24h) across 5 measured providers."
-  - "{{name:monad-official}} answers at {{p50:monad-official}}. On-chain response headers (`x-node-id: monad_monad-mainnet_fra`) confirm geo-distributed nodes co-located with our Railway probes in GCP datacenters (Frankfurt for eu-west, equivalent US and Singapore nodes). The sub-2 ms figure reflects network co-location, not what a developer outside GCP infra will experience — expect 30-120 ms from a typical connection. Rate-limit: 25 rps."
+  - "{{name:monad-official}} answers at {{p50:monad-official}}. On-chain response headers (`x-node-id: monad_monad-mainnet_fra`) confirm geo-distributed nodes co-located with our Railway probes in GCP datacenters (Frankfurt for eu-west, equivalent US and Singapore nodes). The sub-2 ms figure reflects network co-location, not what a developer outside GCP infra will experience; expect 30-120 ms from a typical connection. Rate-limit: 25 rps."
   - "{{name:drpc}} sits at {{p50:drpc}} and {{name:tenderly}} at {{p50:tenderly}}: the multi-chain gateways reached Monad within weeks of mainnet, but route through their global meshes rather than chain-local nodes."
   - "{{name:bloxroute}} at {{p50:bloxroute}} is the outlier entry: a BDN operator exposing a keyless endpoint, unusual for a company whose core product is paid low-latency access."
 

benchmarks/perp-cost-slope.yml

@@ -88,7 +88,7 @@ per_chain_explainer:
 
 findings:
   - "{{name:gains}} and {{name:gmx}} always score 1.00 by construction, not by performance: oracle-priced venues charge a flat percentage of notional with no orderbook, so the $1M all-in is identical to the $1k all-in in basis points regardless of market conditions. Their top leaderboard position reflects fee structure, not execution quality."
-  - "Among orderbook venues, {{name:hyperliquid}} sits at {{p50:hyperliquid}} cost slope (24h avg) — the lowest impact in the cohort at $1M notional, driven by the deepest onchain book."
+  - "Among orderbook venues, {{name:hyperliquid}} sits at {{p50:hyperliquid}} cost slope (24h avg), the lowest impact in the cohort at $1M notional, driven by the deepest onchain book."
   - "Venues showing N/A in the slope column cannot reliably fill a $1M market order. The fill-rate panel shows what fraction of measurement cycles the $1M tier was successfully absorbed."
   - "The cost slope is asset-specific: use the chain tabs to compare ETH (deepest books), BTC (tightest books) and SOL (thinnest books) separately."
 

benchmarks/trading-app-execution.yml

@@ -57,6 +57,24 @@ faq:
   - q: "Which platforms are tracked on EVM chains?"
     a: "GMGN, Maestro, and Banana Gun are tracked on Ethereum. GMGN, Maestro, Axiom, and Banana Gun are tracked on BSC. GMGN, Maestro, and Banana Gun are tracked on Base. pump.fun Terminal is tracked on all three EVM chains."
 
+providers:
+  - slug: axiom
+    name: Axiom
+  - slug: gmgn
+    name: GMGN
+  - slug: trojan
+    name: Trojan
+  - slug: maestro
+    name: Maestro
+  - slug: pump-fun
+    name: pump.fun
+  - slug: photon
+    name: Photon
+  - slug: bullx
+    name: BullX
+  - slug: banana-gun
+    name: Banana Gun
+
 # No prometheus section -- this bench uses a custom page renderer
 # at src/app/benchmarks/trading-app-execution/page.tsx that reads
 # directly from the exec REST API (exec.openchainbench.com).

src/app/apps/page.tsx

@@ -10,6 +10,7 @@ import { TRADING_APPS } from "@/lib/trading-apps-config";
 import { fetchDeFiLlamaData } from "@/lib/defillama";
 import { TradingAppsLeaderboard, type UnifiedAppRow } from "@/components/trading-apps-leaderboard";
 import { RevenueSummary } from "@/components/revenue-summary";
+import { SolanaExecTable } from "@/components/solana-exec-table";
 import Link from "next/link";
 
 const DESCRIPTION =
@@ -38,7 +39,6 @@ export default async function AppsHubPage() {
     fetchDeFiLlamaData(dlSlugMap),
   ]);
 
-  // Market share = each platform's DL 24h fees / cohort total (active only)
   const activeDlTotal = TRADING_APPS.filter((a) => !a.inactive && a.defillamaSlug)
     .reduce((sum, a) => sum + (dlData.get(a.id)?.total24h ?? 0), 0);
 
@@ -69,22 +69,18 @@ export default async function AppsHubPage() {
         const wData = solRow.windows[w];
         if (wData) {
           if (meta.solanaFeeIsUSDC) {
-            // Fees collected in USDC (6 dec): raw units / 1e6 = USD directly
             solanaFees = wData.sumPlatformFeeLamports / 1e6;
           } else if (solPrice !== null) {
             solanaFees = wData.sumPlatformFeeLamports / 1e9 * solPrice;
           }
         }
       }
 
-      // FOMO: supplement on-chain fees with off-chain relay fees from Dune
-      // (relay accounts for ~96% of FOMO's actual revenue)
       if (meta.id === "fomo" && fomoRelay) {
         const relayFee = w === "24h" ? fomoRelay.fees24h : w === "7d" ? fomoRelay.fees7d : fomoRelay.fees30d;
         solanaFees = (solanaFees ?? 0) + relayFee;
       }
 
-      // EVM: 24h adds native ETH/BNB in USD; 7d/30d stable only (no historical prices).
       const evmStable = (chain: typeof ethChain) =>
         !chain ? null : w === "24h" ? chain.stable24h : w === "7d" ? chain.stable7d : chain.stable30d;
       const evmNative = (chain: typeof ethChain) =>
@@ -165,6 +161,32 @@ export default async function AppsHubPage() {
         </div>
       )}
 
+      <div className="mt-10 border-t border-rule pt-8">
+        <h2 className="text-lg font-semibold text-ink mb-1">Solana execution quality</h2>
+        <p className="text-sm text-ink-soft mb-6">
+          Priority fees, Jito bundle rates, and platform fees per transaction — measured passively from on-chain data.
+        </p>
+        <SolanaExecTable
+          platforms={solanaData?.platforms ?? []}
+          updatedAt={solanaData?.updatedAt ?? null}
+        />
+        <div className="mt-4 text-xs text-ink-muted leading-relaxed max-w-2xl space-y-1.5">
+          <p>
+            <strong>Priority fee</strong> = total tx fee minus the 5 000-lamport base fee per signature.
+          </p>
+          <p>
+            <strong>CU price</strong> = priority fee ÷ compute units consumed (microlamports/CU). p50/p95 shows median and tail pressure.
+          </p>
+          <p>
+            <strong>Jito rate</strong> = fraction of transactions tipping a Jito tip account. Higher = more MEV-sensitive routing.
+          </p>
+          <p>
+            <strong>Platform fee</strong> = avg SOL to the platform fee account per transaction.
+            Source: Helius enhanced transaction API. No synthetic trades, no on-chain footprint.
+          </p>
+        </div>
+      </div>
+
       <div className="mt-10 border-t border-rule pt-8">
         <p className="text-xs font-medium text-ink-muted uppercase tracking-wide mb-3">Related benchmarks</p>
         <div className="flex flex-wrap gap-3">
@@ -187,8 +209,8 @@ export default async function AppsHubPage() {
           >
             <span className="text-lg">⚡</span>
             <div>
-              <p className="text-sm font-medium text-ink group-hover:text-accent transition-colors">Execution Quality</p>
-              <p className="text-xs text-ink-muted">Priority fees, Jito rates, platform fees</p>
+              <p className="text-sm font-medium text-ink group-hover:text-accent transition-colors">Execution Quality Bench</p>
+              <p className="text-xs text-ink-muted">Priority fees, Jito rates, platform fees — full dataset</p>
             </div>
             <svg className="ml-auto text-ink-faint group-hover:text-ink-muted transition-colors" width="14" height="14" viewBox="0 0 12 12" fill="none" aria-hidden="true">
               <path d="M3.5 3H2a1 1 0 00-1 1v6a1 1 0 001 1h6a1 1 0 001-1V8.5M7 1h4m0 0v4m0-4L5.5 6.5" stroke="currentColor" strokeWidth="1.3" strokeLinecap="round" strokeLinejoin="round"/>

src/components/exec-bench-table.tsx

@@ -4,6 +4,7 @@ import { useState } from "react";
 import type { ExecLeaderboardResponse } from "@/lib/solana-exec";
 import { PLATFORM_DISPLAY, lamportsToSOL, fmtCUPrice } from "@/lib/solana-exec";
 
+const NOW_MS = Date.now();
 const WINDOWS = ["24h", "7d", "30d"] as const;
 type Window = (typeof WINDOWS)[number];
 
@@ -28,6 +29,10 @@ export function ExecBenchTable({
   data: ExecLeaderboardResponse | null;
 }) {
   const [win, setWin] = useState<Window>("24h");
+  const stale =
+    data?.updatedAt
+      ? NOW_MS - new Date(data.updatedAt).getTime() > 3 * 60 * 60 * 1000
+      : false;
 
   if (!data) {
     return (
@@ -37,16 +42,11 @@ export function ExecBenchTable({
     );
   }
 
-  const sorted = [...data.platforms].sort(
+  const sorted = [...(data.platforms ?? [])].sort(
     (a, b) =>
       (b.windows[win]?.txCount ?? 0) - (a.windows[win]?.txCount ?? 0),
   );
 
-  const stale =
-    data.updatedAt
-      ? Date.now() - new Date(data.updatedAt).getTime() > 3 * 60 * 60 * 1000
-      : false;
-
   return (
     <div className="card rounded-xl overflow-hidden">
       <div className="px-4 sm:px-6 pt-4 pb-0 flex items-end justify-between gap-4 flex-wrap border-b border-rule">