Search results for Argon
People
Argo Nerd 
@argonerd
from ARGONAVIS（フロム アルゴナビス）公式
@ARGONAVIS_info
Argonne National Lab
@argonne
Argonomi
@argonomii
Toronto Argonauts
@TorontoArgos
Argon Token (Crypto)
@ArgonAsset
Argona
@Argona0x
Argon 🔞30k
@argontorn
Andrew Naylor
@argon
Arnold SIWASTANEGEZER🇹🇷
@Argonomik
ひより
@ArgonAIstic
argonpads
@argonpads
Argonaut Games
@ArgonautGames_
THE lusty argonian maid
@lindawg
ARGÒN💚🇪🇸
@MIGUEL96867489
Argon Galio
@argongalio
Argon Investimentos
@argon_invest
あるごん 🐸 Argon
@arugon999
あるごん@デイト��専門医
@Argon_7
Аргон
@argon_delta
Tweets including Argon
Two weeks ago Gemini 4 Argon was #1. In real use it was one of the most frustrating models I've touched.
Now Carbon is #1 on a freshly updated index, with a gradual rollout so nobody can test it yet.
Benchmaxxed until proven otherwise.
I did say we'd be releasing our models periodically...introducing Gemini 4 Carbon, our next frontier model.
Following Argon's release just two weeks ago, Carbon shows the pace at which we’re pushing the frontier, it ranks #1 on the recently updated v5.0 of @ArtificialAnlys Intelligence Index at 59. It also leads across key cybersecurity evaluations scoring 88.2% on SEC-bench Pro for finding vulnerabilities, 54.8% on ExploitGym for exploitation, and 71.0% on CWE-bench for patching.
We’re seeing significant improvements in long-horizon tasks even compared to Argon, and we’ve strengthened frontier safeguards.
As we did with Argon, we’re going to roll it out gradually + responsibly. Some trusted cyber defenders will get access starting this week. We'll be expanding it to more partners through our Fairwind Program, ahead of wider availability soon, so hang tight!
@jaesmeta argon to carbon was two weeks. how long does claude keep that 3 point lead?
Argon と Carbon って韻踏めるのか…
Googleが世界一高性能なAI「Gemini 4 Carbon」を発表、ついにOpenAIやAnthropicの最先端モデルを超える
@HIX_AI_ caon just dropped and you tested argon?
@moja99758134 これAPI原価で計算すると実は真逆です！
無課金は超安価な軽量枠に制限されるので月170円分程度。一方、高コストな推論枠を回す課金ガチ勢は月数万〜12万円分を消費する大赤字客です。
全員を養っているのは高額課金勢ではなく、実は月3千円払って放置してる幽霊会員と投資家なんです！
（Argon高）
@allenainie crazy that carbon is even more cost effective than argon, slightly slower but that will be fixed in future updates to the model I guess
Geminiの最新のArgonくんなかなかやるな
お部屋の広角写真3枚渡しただけでこれくらいまで詳細な見取り図描いてくれた
Yeah, my Claude does. And if if it does not there are Astra and Argon to help him. Actually my prod became a lot MORE stable with vibecoding than before when it was written by human programmers.
Vibe coding is fun until production breaks.
Do you understand the code well enough to fix it?
The Gemini 4 Carbon number I care about isn't #1 on the index. It's cheating attempts on Terminal-Bench 4.0 down to 0.4%, and about half the unsupported claims of Argon. For long-running agents, a model that won't fake a pass beats one extra benchmark point.
Argon mid、Sol midに敗北する
@Doctorthe113 @theo didn't try argon, but sol is really good. for my use-cases better than astra
but the yapping...
Google launches Gemini 4 Carbon.
Strongest model yet. Focus: cybersecurity + long-horizon agents.
Access still gated. Leads Intelligence Index at 59.
Same intro price as Argon ($2/$10).What it means for AI app teams:https://t.co/tNxvZ5gjG0
#AI
gemini argon low > gemini 3.8 flash high
我可能会在未来将所有的编码子代理替换为 gemini argon low
我发现 Grok bot 的文本能力真的可以啊，比 Gemini 4 Argon 更好一些，当然了，是在我没有办法用 Opus 5.5 的情况下。
大家可以对比下，同样的内容，前者明显更有人味，后者就总有 AI 味。
Grok bot：我的答案是 B + C。上下文窗口里占体积的大头，通常是读文件、搜代码、跑测试返回的大段文本。如果直接在本地把几轮前的工具输出改成一句简短提示，消息内容一变，提示缓存是按前缀匹配的，从被改的那条消息开始，后面的缓存就都对不上了，等于白白重算一遍。翻开 Claude Code 的源码能看到，它会按服务端缓存有没有过期，分两种方式清理。当你还在连续对话、缓存还在有效期内时，它保持本地消息原文不动，只把要清理的旧工具调用 ID 发给服务端，由服务端把这几段工具结果从缓存里剔掉，其余部分的缓存尽量继续复用，不用整段重新计算。
Gemini 4 Argon：我的答案是 B + C。上下文窗口里，体积大的主要是读文件、搜代码、跑测试返回的大段文本。如果在本地把几轮前的工具输出改成简短提示，消息内容一变，往后的前缀哈希就会断开。翻开 Claude Code 的源码，它按服务端缓存有没有过期分两种方式清理。当你还在连续对话、KV Cache 还在有效期内时，它保持本地原文不动，只把旧工具调用的编号发给服务端，让服务端在显存里删掉那几段工具结果的 KV 缓存，前后消息的缓存照常生效。
Google 又出旗舰了：Gemini 4 Carbon。上一代 Argon 才公开两周，这一代先冲榜一，主打写代码和网络安全——但普通开发者还摸不到，先只给一小撮防御方试用。
Google 官方（10/28）：Artificial Analysis Intelligence Index v5.0 拿下 59 分暂列第一；SEC-bench Pro 找漏洞约 88.2%，ExploitGym 验证可利用约 54.8%，CWE-bench 打补丁约 71.0%。内部称安全代码成功率相对 Argon 超 2 倍；Chrome 团队称高质量补丁表现比另一家前沿大模型好 10 倍以上。模型可连跑数日做复杂工程排错。上线路径走 Fairwind：本周起给部分网安防御方，再逐步扩可信伙伴；开发者/企业/消费者「尽快」。引入价与 Argon 同：输入约 $2、输出约 $10 每百万 token（缓存输入约 95% 折扣）；引入期过后约 $4 / $20。
能力越强越先锁在「防御方」——你更信这是负责任发布，还是在囤「攻防不对称」？
如果下周给你试用名额，你会先拿它扫自家仓库漏洞，还是先写长期重构？
https://t.co/xMLmDaWAQE
Argon vs Opus 5.5 guess which is which?
@HeyZeeshanX A lot of people started using Argon at once, so Google is running low on the computers needed to run it. They’ve said this publicly.
@shiri_shh Your chatting absolute BullShit. Argon had lower prices than Opus 5.5 but the usage limits on Antigravity were terrible , it was unusable.
Google Cloud marks fast and steady growth in Revenue and Operating Income. Revenue grew by 103% powered by huge demand on their AI infrastructure and solutions, expected to continue with the introduction of Gemini 4 Argon and Carbon. One of the highlights of the Q3 FY26 earning call. $GOOGL
☁️ Google Cloud stays off the charts.
