jinllll996

[jinllll996 @noopepty]

@dingyi 我在codex运行几十个小时没停的并行goal都不卡，非常流畅。cursor倒是会非常卡。64g m3max

[jinllll996 @noopepty]

@achao_talks @dotey @LinearUncle 只是你看到的隔天，那个issue上周就有了

[jinllll996 @noopepty]

@imyouhu 那你啥也别干了最好的mythos不给你用

[jinllll996 @noopepty]

@llxth1 @TaNGSoFT https://t.co/rgpF6dNDWW

https://t.co/C6lhsLORx2

就只刷中推呗

[Philipp Schmid @_philschmid]

Interesting new SWE/agentic benchmark (DeepSWE) was released yesterday. 113 tasks across 91 repos in 5 languages. Here are interesting things I noticed:

- The evaluation harness (mini-swe-agent) gives every model a single bash tool and the same SI. No vendor editing primitives.

- Eval Prompts are shorter than SWE-Bench Pro, but require 5.5× more code and touch 7 files on average. The idea is to mimic how developers actually talk to agents, short behavioral descriptions, not verbose specs.

- SI describes a specific workflow: find code, reproduce, fix, verify, edge cases, submit. This maps directly onto how the verifier grades, which could bias toward models that follow instructions literally over models that explore more.

- The bash tool is guarded, outputs over 10k chars get truncated. Malformed tool calls get caught and retried with guidance rather than crashing. To prevent to blow up context.

- Mini-swe-agent claims to match or beat 1P harnesses on the same tasks. Claude Opus scored +10pp over Claude Code. Gemini 3.1 Pro scored +20pp over Gemini CLI.

Would love to see how other harness × model combinations will do, e.g. @cursor_ai, @antigravity, @FactoryAI and how well the eval harness does on more general knowledge work, e.g. GDPval.

Great to see the SWE-agent team keep pushing on both the research and eval side. 🤗

[jinllll996 @noopepty]

@jesselaunz 证明了简介里写AI的推的平均水平能有多水

[jinllll996 @noopepty]

@0xshimei cache比例都没给你，谈价格你不纯扯淡呢？
36亿 token也就半个5x pro的用量

[jinllll996 @noopepty]

@corndogjpn2 傻逼40克蛋白质给你吓傻了？

[jinllll996 @noopepty]

@joshlovejoe @qingyue820 opus4.7连gpt5.2都不如 你会写代码吗🤡？

[jinllll996 @noopepty]

@thaddeusjiangzh 阅读理解满分

[jinllll996 @noopepty]

@supezen 前15秒每秒1%，后面越来越慢，最高95%
完全写死的算法

[jinllll996 @noopepty]

@wey_gu 先后顺序搞错了，先有的codex app

[jinllll996 @noopepty]

@ZaynHao 跟openai比产品能力？脑子坏了？

[jinllll996 @noopepty]

@AlchainHust 算了吧 opus4.7写代码连5.2都赶不上

[jinllll996 @noopepty]

@Lonely__MH 吃都吃不上几口那你吃你妈逼的细糠呢弱智

[jinllll996 @noopepty]

@FuSheng_0306 因为你只会写UI

[jinllll996 @noopepty]

@mtrainier2020 说明你用不明白codex

[jinllll996 @noopepty]

@JackyFuxx @CEOBriefing 傻逼用过几天claude啊？codex周活马上5M了Anthropic除了claude code cli端还有像样的c端产品？不尊重工程师 自大狂当ceo的傻逼公司还有你这种孝子呢？

[jinllll996 @noopepty]

@tualatrix goal achieved*
ralph* loop
可以看出是自己好好打字了😂