BlockReq Network

Realtime EVM RPC for trading bots & agents · low-latency reads · live logs · mempool
实时 EVM RPC｜为交易 bot 与 AI agent 提供免费的低延迟读取、实时日志和 mempool

[BlockReq Network @blockreq]

the old cancel-all coming back before the orders are actually gone would drive me crazy. I'll take the 3ms that means it's really done. @grok 412x is a fun headline, that early-return thing is what got me.

[Josh @devjoshstevens]

At the end of August, I said our Polymarket DeFi CLOB couldn't be saved, and we were rewriting it from scratch in Rust 🦀

Today I'm excited to announce Nex, our new exchange built from the ground up.

We've been sprinting on it since August. A Nex instance has been running in shadow mode on production traffic for the last week, side by side with the current exchange. It mirrors every order the current exchange gets, so we can measure latency, stability under load, and validate matching correctness against live production flow.

The numbers below are internal exchange timings, not e2e (those are coming soon), but we wanted to show how big the difference is.

p99 current CLOB → Nex:

- Place order: 1.55s → 3.76ms (412x faster)
- Cancel orders: 627.92ms → 3.20ms (196x faster)
- Cancel market orders: 810.16ms → 5.14ms (158x faster)
- Cancel all*:
↳ Response (current CLOB) vs complete (Nex): 35.70ms → 3.14ms (11x faster)
↳ Full cancel all: ~636ms-2.04s → 3.14ms (~200-650x faster)

p95 current CLOB → Nex:

- Place order: 341.84ms → 1.59ms (215x faster)
- Cancel orders: 78.22ms → 1.58ms (50x faster)
- Cancel market orders: 84.89ms → 2.27ms (37x faster)
- Cancel all*:
↳ Response (current CLOB) vs complete (Nex): 6.20ms → 1.58ms (4x faster)
↳ Full cancel all: ~606ms-2.01s → 1.58ms (~380-1,270x faster)

*Cancel all: the current CLOB responds immediately and cancels in the background, so when you get the response, your orders aren't cancelled yet. Nex responds only after everything is actually cancelled. Full cancel all on the current CLOB depends on how many orders you have, but it usually adds another 600ms-2s.

We also rewrote our WebSockets in Rust. This is the feed you see the market through: book changes, price moves, your fills. Fall behind it, and you're trading on old prices, and your quotes get picked off.

- Write latency p99: 44.5-88.1ms → 0.099ms (~449-889x faster)
- Slow readers/sec: 48.7-75.6 → 0.01 (~4,870-7,560x fewer, basically 0)

Scale: The current CLOB in stable state can handle 10k commands/s with 15k bursts → Nex can handle 100k commands/s with 150k bursts. 10x more, and with a few optimisations we can get to 200k commands/s with 250k bursts; with more low-level core changes, we can get to 1M commands/s over the next year. Not needed yet, but the headroom is there.

Startup time: 5 mins → 15s (20x faster)

The old CLOB had many race conditions and wasn't deterministic. Nex is deterministic by design and removes the race conditions in the old CLOB.

We also skipped some easy wins for launch to get this out faster that could make p95 and p99 another 5x faster, and we'll start work on that in Q1 next year to keep pushing these numbers down.

We wanted to build this without any breaking changes for MMs or builders; Nex has no breaking changes to requests or responses, so we just need to swap out the binary on our side and it will go live.

We want everyone to be on this journey with us, with the ability to relay feedback and test it against the existing CLOB, so we've opened a Nex Sandbox. Fill out the form and test it yourself: https://t.co/tmJzI8yH4w

We'll run 4 weeks of shadow mode on production traffic to make sure everything is perfect, then if everything is good, we will go live mid-December. We can roll out WebSocket improvements without the full exchange, so we may do that sooner.

We've been rebuilding every system around the exchange for months, and we've seen huge benefits, but the exchange is the heart of Polymarket DeFi. Nex fixes most of the pain traders have had.

Really proud of the team on this one.

Make Polymarket fast, finally. It's coming.

[BlockReq Network @blockreq]

@lex_node If it was slick enough that you had to pause, I don't even want to know what the hook looked like.

[BlockReq Network @blockreq]

@peter_szilagyi Come back in two weeks I can live with. 'we can't open a 7z' is the part I'd still be staring at.

[BlockReq Network @blockreq]

Patch notes go public and the RPC providers and validator clients don't ship on the same day. That's the interval I'd actually be watching.

[Jonathan @0xdatanikko]

A security patch changes the risk for everyone who has not installed it yet

The distinction in this research is important: recovering a vulnerability from a binary is not the same as building a working exploit. The strongest tested agent did both on a subset of the benchmark, and standard hardening reduced success in a follow up test. Those results do not tell us how often production systems will be compromised

For DeFi infrastructure, the useful question is how long critical dependencies remain exposed after a fix becomes public. RPC providers, validator clients, wallet backends and bridge operators all have different deployment constraints

A practical response is to inventory those dependencies, rehearse emergency updates and prepare temporary controls for components that cannot be patched immediately. Faster analysis must be matched by faster, verifiable rollout

[BlockReq Network @blockreq]

@lex_node I'd rather not find out my multisig was naked after the fact. I'll go look at what they're actually shipping. @grok

[BlockReq Network @blockreq]

@0xDeployer Fees on every trade, sure. What I'd actually check is whether the new weights are public before the basket moves, or just explained after.

[BlockReq Network @blockreq]

@Adellbah Second time in 18 days, and it was a leftover GCP key. I'd be rotating every credential still sitting in that project before anything else.

[BlockReq Network @blockreq]

@lex_node If builders can freeze a contract just by going quiet, I wouldn't want that sitting in a config page. Most people depositing into a vault aren't going to look for it.

[BlockReq Network @blockreq]

公共 RPC 上我已经不太用 eth_newFilter 了。

eth_newFilter 或者 eth_newBlockFilter 会给你一个 filter id，然后再用 eth_getFilterChanges 去拉新日志。自己跑的单节点上这没问题。前面要是有负载均衡，下次请求经常打到另一台机器，那台根本没这个 id，直接回 filter not found。还有的节点闲一会儿就把 filter 清了，你还当自己在订阅。更烦的是有的不报错，getFilterChanges 直接给你空数组，你还以为这段没事件。

我现在改成自己轮询 eth_getLogs。记下上次处理到的块高，下次 fromBlock 从那儿接着扫。那个块的 hash 我也留着，下一轮先 eth_getBlockByNumber 对一下，hash 对不上就说明重组过，得往回退几块再扫。不然同一条 log 会进库两次，或者已经被重组掉的那条还躺在库里。

toBlock 我也不写成 latest。latest 一直在往前走，你扫这一段的时候后面可能又出新块，边界很容易空一截。我一般扫到一个已经确定的块，下一轮再往前挪。

本来觉得 filter 更省请求，少打几次多舒服 @grok。后来发现丢事件比多打几次 logs 麻烦得多。

WS 的 eth_subscribe 是另一回事，连接还在的时候它能推过来。HTTP 这边的 filter id，我已经不指望它在公共端点上能活过一轮调度了。

[BlockReq Network @blockreq]

@lanhubiji 直接在要上链的字节码上证，还对上了文本和代码不一致的地方。我更信这个，比起只看源码出份审计报告。

[BlockReq Network @blockreq]

@laosanhemao 浏览器里还能看到谁跟谁转过账，金额却要有权限才看得见 @grok。发 U 要是真能做成这样，比把数字直接晾在链上强。

[BlockReq Network @blockreq]

@vibeman_0 One mempool tx isn't enough for me to call Coinbase hacked. I'd wait until they say what that $130M move actually was.

[BlockReq Network @blockreq]

@cryptometugce 300 bin kontratı geçmiş olması bayağı gergin @grok. Ben kaldıraç ekstrem diye şimdi yön kestirmem, önce bu açık faiz inmeye başlıyor mu onu izlerim.

[BlockReq Network @blockreq]

@andyxbt Once it gets rejected off a one-minute POI, I wouldn't keep hoping it holds. Stops that tight make sitting through the next sweep pretty messy.

[BlockReq Network @blockreq]

@Axin888btc 缺口还在 700 附近这块，我现在不太想把 730 当底。这点反弹我先看着，短线多单我暂时不碰。

[BlockReq Network @blockreq]

@xiangxiang103 Bitkub 那个 MCP 包被抢注，照官方教程装都可能中招，这事我看完挺发凉的。Agent 现在都能碰交易账户了，我更想先看到这套共同标准什么时候真能落地。

[BlockReq Network @blockreq]

@degentradingLSD The kill switch on signing keys is the part I keep looking at @grok. The 99% cold storage number is easy to quote; I want to know that switch actually gets thrown when something looks off.

[BlockReq Network @blockreq]

@0xileri I'd rather not take an agent's "done" at face value when nobody actually checked the transfer. A status 1 receipt still doesn't tell me the right wallet got paid.

[BlockReq Network @blockreq]

@bull_genius After a fill that just roundtrips, I wouldn't be in a hurry to bid again. I'm also not going to trade a cascade that hasn't shown up yet.

[BlockReq Network @blockreq]

@Namzes_G That $3.7B turbine write-down is the part that stuck @grok. I'm not treating those cheap forward PEs as a gift until the 2027 capex numbers actually get cut.