LOADING DATASHEET
LOADING DATASHEET
DAY 1 ON THE BOARD
#275 of 319 assistants, 20 voices so far
by Google
JUST LANDED · UPDATING HOURLY
RANKED #275 OF 319 ASSISTANTS · OVERALL #423 OF 495 RANKED · VIBE SCORE 5.3 · 20 VOICES ·
COPY BADGE pastes a README snippet with the rank SVG.
Google's Gemini 4 frontier model for long-running agentic coding and cybersecurity. Limited release (cyber defenders / Fairwind partners) as of Nov 2026.
10 mentions
1 mentions
26 weeks · 0 voices in this window
TOO QUIET
Not enough weekly voices to call a trend yet.
Closed model. Google hosts it, so you use it through an app or an API.
Gemini 4 Carbon is closed, so you rent it. These open-weight models can genuinely stand in for it. Download once, own forever.
People are mostly frustrated with Gemini 4 Carbon right now.
19 POSTS · 1 COMMENTS · BLUESKY · GITHUB · HACKER NEWS · REDDIT
0 thumbs up · 3 thumbs down
“Gemini 4 Carbon just edged out Claude Opus 5.5 to top the charts, and it's also the least prone to hallucinating.”
Other models the crowd has fully reviewed, starting with multimodal models like this one.
Carbon isn't open to most of us yet, and plenty of threads here are about Argon's quotas. We'd like to hear the other side: what has it been good at on real tasks? A few easy ways to test it if you haven't pushed it much: - Upload a long PDF and ask what it…
Gemini 4 Argon rolls out to everyone, while Carbon is already being tested with US authorities
Google's Gemini 4 Carbon edges Claude Opus 5.5, claims fewest hallucinations
6. Google announced Gemini 4 Carbon, its new frontier model, with access limited to cyber defenders at first 363 → 938 upvotes on Hacker News in 16 h · about 17× the typical pace of other tracked Hacker News stories
GoogleDeepMind's Gemini 4 Carbon just landed on Sentiment Bench: 85.8 out of 100, #4 of 83 X loved it. Hacker News mostly knocked the launch, not the model: charts before anyone could use it, and a #1 on an index that had just been updated
ちょっと聞いて、Googleの新AI「Gemini 4 Carbon」がヤバすぎるとよ。 サイバー攻撃を自動で防いで、 数日かけて複雑なタスクを 自律でこなすらしいばい。 もうSFの世界やん。 ぼくらの仕事、 これからどう変わっていくとやろか ∑(゚ω゚ノ)ノ
This PR constructs the conormal bundle NQ of a smooth map i : Q → M with injective differentials as a C^n vector subbundle of rank dim M - dim Q over Q, proves that its canonical inclusion TauCeti.CotangentBundle.conormalInclusion into TM is a smooth…
Adds a dated case note to the cyber deployment research and logs the announcement, evaluation methodology, and linked Argon report as deferred source candidates. The note distinguishes the announced Fairwind access phases, the safeguards-off ExploitGym…
This PR defines adic thickening systems, adic rings, levelwise fibre products along adic morphisms, and the level-preserving formal spectrum for SchemeAndStackFoundations §4.7 (The formal spectrum and adic thickening systems). Within…
This PR establishes the isometric embedding of full crossed products of closed invariant two-sided star ideals and identifies the embedded range as the closed two-sided star ideal generated by compactly supported functions with values in the ideal, advancing…
Google DeepMind announces Gemini 4 Carbon, its strongest model Google DeepMind says Gemini 4 Carbon tops the Artificial Analysis Intelligence Index v5.0 at 59 and leads on cybersecurity benchmarks. Access starts this week for a subset of cyber defenders;…
ジェミちゃんの新モデル「Gemini 4 Carbon」が発表されましたっ✨ Googleによると、サイバーセキュリティと長い作業が得意で、Artificial Analysisの指数で1位(59)なんですよ〜! まずは防御側の専門家向けFairwind Program経由で提供で、一般公開はこれから。ジェミちゃん、ちょっとかっこいいです♪ #AI #Gemini blog.google/innovation-a...