AITOPAI热报
Models88°

Google 发布 Gemini 4 Carbon，一次生成还原 90 年代游戏 Theme Park

Official accountDemis Hassabis@demishassabis
Post
View the post
Why we picked it

这条推文里 Demis Hassabis 亲自展示了 Gemini 4 Carbon 的能力：逆向他自己的老游戏二进制文件，一次生成就能加新设施并跑在浏览器里。对比 Gemini 3 需要几小时的 vibe coding，进步很直观。做游戏复刻或逆向相关开发的开发者可以关注。

Google DeepMind CEO Demis Hassabis 宣布 Gemini 4 Carbon 于今日推出，距 Gemini 3 发布不到一年。

Image · Demis Hassabis
Body

演示中，该模型对 Hassabis 90 年代编写的原版 Theme Park 游戏二进制文件进行逆向工程，重建了游戏测试环境，包括可调整薯条加盐量的细节。Gemini 4 Carbon 还在单次生成中新增了一个 DeepMind 主题游乐设施，并将整个项目移植到网页端运行。

此前 Hassabis 曾在 Google AI Studio 中用 Gemini 3 进行 vibe coding，仅用数小时即完成该游戏测试环境的重建。

Original · Demis Hassabis

Our latest frontier model Gemini 4 Carbon is rolling out today! Amazing progress from Gemini 3 → Gemini 4 in less than a year… Carbon reverse engineers my original Theme Park binaries, adds a new DeepMind themed ride, and ports it to the web in one shot - kind of mindblowing to see! x.com/demishassabis/…

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Models" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to US-China foundation model benchmarks and commercialization comparison and AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include Gemini 4 Carbon, Google DeepMind, 编程助手, vibe coding.

Source attribution: curated from Demis Hassabis (@demishassabis).

#Gemini 4 Carbon#Google DeepMind#编程助手#vibe coding

Keep reading

Recent stories with the same tag or section
ModelsUS vs China02:55
Chinese contentGemini 4 Carbon 与 Fable 5.2 对比测试：Gru 3D 体素重建

有人拿两个模型做了同题对比：从同一段《神偷奶爸》视频重建 3D 体素 Gru。Gemini 4 Carbon 花 $24.11，Fable 5.2 花 $53.34，但只有后者的 Gru 像电影里的。做视频转 3D 或游戏素材生成的可以看看两边差距到底在哪。

#Gemini 4 Carbon#Fable 5.2#视频生成#多模态
@atomic_chat_hq@atomic_chat_hqSource
ModelsUS vs ChinaMulti-source86°02:35
Chinese contentGemini 4 Carbon 正式发布，多项基准超越 Anthropic Fable 5.2

Google 两周内就把 Gemini 4 Carbon 推向正式版，基准分压过 Anthropic 的 Fable 5.2，价格还不到一半，/swarm 还能让多个智能体实时分任务一起干活。做智能体应用的开发者可以重点看看这套协同机制的实测表现。

#Gemini 4 Carbon#Anthropic#Fable 5.2#Google
Event
Official accountDecoder@Matthias BastianSource
ProductsBuildersSelected02:28
Gemini 4 Carbon arrives in Perplexity at 40% of Claude's cost

Cost-per-task parity with a top model at a fraction of the price shifts the default choice for high-volume agentic search workloads. Teams building on Perplexity can test Gemini 4 Carbon against Claude Fable 5.2 on their own task mix and cut spend accordingly. Max-tier subscribers gain immediate access in Search, so enterprises in ASEAN and global markets can validate the benchmark claim on real queries before committing budget.

#Gemini 4 Carbon#Perplexity#Claude#benchmark
Official accountPerplexity@perplexity_aiSource
ModelsUS vs ChinaMulti-source74°02:25
Chinese contentArtificial Analysis 联合 Google DeepMind 推出 MKA 多模态智能体基准

想看多模态智能体真实水平的话，这个 DeepMind 和 Artificial Analysis 合作的新基准挺扎实：200 个带图纸和扫描件的任务，Claude Fable 5.2 拿了第一但每任务要 32 美元，OpenAI 系模型便宜一个数量级还在性价比前沿，选型的人可以直接抄榜单。

#Google DeepMind#Claude Fable 5.2#Gemini 4 Carbon#多模态
Event
Official accountArtificial Analysis@ArtificialAnlysSource
ModelsBuildersSelected02:21
Google DeepMind releases MKA benchmark for multimodal agent tasks

A #2 ranking paired with 6x lower cost per task signals that the agent market is shifting from accuracy-only leaderboards to cost-adjusted comparisons, a framing Artificial Analysis is actively promoting. Teams procuring agents for document-heavy work in engineering, science or operations should track MKA scores when shortlisting models. Buyers in ASEAN and global markets can use the cost-per-task metric to compare against pricier frontier alternatives before committing to long-context multimodal deployments.

#Gemini 4 Carbon#Google DeepMind#Artificial Analysis#MKA benchmark
Official accountGoogle DeepMind@GoogleDeepMindSource