#AlphaEvolve

[interney]

AlphaEvolve: o que é e o que revela sobre o futuro da inovação nos negócios

Se por muito tempo os sistemas de inteligência artificial (IA) se limitavam a responder perguntas, automatizar tarefas ou reproduzir padrões conhecidos, 2025 marca um novo ciclo. Estamos diante de modelos de IA que não apenas replicam conhecimento, mas que têm potencial real para gerar descobertas inéditas, inclusive aquelas que nem mesmo especialistas humanos haviam identificado. É o caso do…

[dishdasha]

Oh look I know that this will save money and scarce resources, and it's AI that is actually doing something meaningful and isn't costing a human a job somewhere else but... I feel really deflated by this news.

I feel like inventing a machine to do all the inventing for us is like - giving up? Isn't one of the wonderful things about human output, the guts and glory of the human endeavour that leads to it?

[govindhtech]

AlphaEvolve Coding Agent using LLM Algorithmic Innovation

AlphaEvolve

Large language models drive AlphaEvolve, a powerful coding agent that discovers and optimises difficult algorithms. It solves both complex and simple mathematical and computational issues.

AlphaEvolve combines automated assessors' rigour with LLMs' creativity. This combination lets it validate solutions and impartially assess their quality and correctness. AlphaEvolve uses evolution to refine its best ideas. It coordinates an autonomous pipeline that queries LLMs and calculates to develop algorithms for user-specified goals. An evolutionary method improves automated evaluation metrics scores by building programs.

Human users define the goal, set assessment requirements, and provide an initial solution or code skeleton. The user must provide a way, usually a function, to automatically evaluate produced solutions by mapping them to scalar metrics to be maximised. AlphaEvolve lets users annotate code blocks in a codebase that the system will build. As a skeleton, the remaining code lets you evaluate the developed parts. Though simple, the initial program must be complete.

AlphaEvolve can evolve a search algorithm, the solution, or a function that creates the solution. These methods may help depending on the situation.

AlphaEvolve's key components are:

AlphaEvolve uses cutting-edge LLMs like Gemini 2.0 Flash and Gemini 2.0 Pro. Gemini Pro offers deep and insightful suggestions, while Gemini Flash's efficiency maximises the exploration of many topics. This ensemble technique balances throughput and solution quality. The major job of LLMs is to assess present solutions and recommend improvements. AlphaEvolve's performance is improved with powerful LLMs despite being model-agnostic. LLMs either generate whole code blocks for brief or completely changed code or diff-style code adjustments for focused updates.

Prompt Sample:

This section pulls programs from the Program database to build LLM prompts. Equations, code samples, relevant literature, human-written directions, stochastic formatting, and displayed evaluation results can enhance prompts. Another method is meta-prompt evolution, where the LLM suggests prompts.

Pool of Evaluators

This runs and evaluates proposed programs using user-provided automatic evaluation metrics. These measures assess solution quality objectively. AlphaEvolve may evaluate answers on progressively complicated scenarios in cascades to quickly eliminate less promising examples. It also provides LLM-generated feedback on desirable features that measurements cannot measure. Parallel evaluation speeds up the process. AlphaEvolve optimises multiple metrics. AlphaEvolve can only solve problems with machine-grade solutions, but its automated assessment prevents LLM hallucinations.

The program database stores created solutions and examination results. It uses an evolutionary algorithm inspired by island models and MAP-elites to manage the pool of solutions and choose models for future generations to balance exploration and exploitation.

Distributed Pipeline:

AlphaEvolve is an asynchronous computing pipeline developed in Python using asyncio. This pipeline with a controller, LLM samplers, and assessment nodes is tailored for throughput to produce and evaluate more ideas within a budget.

AlphaEvolve has excelled in several fields:

It improved hardware, data centres, and AI training across Google's computing ecosystem.

AlphaEvolve recovers 0.7% of Google's worldwide computer resources using its Borg cluster management system heuristic. This in-production solution's performance and human-readable code improve interpretability, debuggability, predictability, and deployment.

It suggested recreating a critical arithmetic circuit in Google's Tensor Processing Units (TPUs) in Verilog, removing unnecessary bits, and putting it into a future TPU. AlphaEvolve can aid with hardware design by suggesting improvements to popular hardware languages.

It sped up a fundamental kernel in Gemini's architecture by 23% and reduced training time by 1% by finding better ways to partition massive matrix multiplication operations, increasing AI performance and research. Thus, kernel optimisation engineering time was considerably reduced. This is the first time Gemini optimised its training technique with AlphaEvolve.

AlphaEvolve optimises low-level GPU operations to speed up Transformer FlashAttention kernel implementation by 32.5%. It can optimise compiler Intermediate Representations (IRs), indicating promise for incorporating AlphaEvolve into the compiler workflow or adding these optimisations to current compilers.

AlphaEvolve developed breakthrough gradient-based optimisation processes that led to novel matrix multiplication algorithms in mathematics and algorithm discovery. It enhanced Strassen's 1969 approach by multiplying 4x4 complex-valued matrices with 48 scalar multiplications. AlphaEvolve matched or outperformed best solutions for many matrix multiplication methods.

When applied to over 50 open mathematics problems, AlphaEvolve enhanced best-known solutions in 20% and rediscovered state-of-the-art solutions in 75%. It improved the kissing number problem by finding a configuration that set a new lower bound in 11 dimensions. Additionally, it improved bounds on packing difficulties, Erdős's minimum overlap problem, uncertainty principles, and autocorrelation inequalities. These results were often achieved by AlphaEvolve using problem-specific heuristic search strategies.

AlphaEvolve outperforms FunSearch due to its capacity to evolve across codebases, support for many metrics, and use of frontier LLMs with rich context. It differs from evolutionary programming by automating evolution operator creation via LLMs. It improves artificial intelligence mathematics and science by superoptimizing code.

One limitation of AlphaEvolve is that it requires automated evaluation problems. Manual experimentation is not among its capabilities. LLM evaluation is possible but not the major focus.

AlphaEvolve should improve as LLMs code better. Google is exploring a wider access program and an Early Access Program for academics. AlphaEvolve's broad scope suggests game-changing uses in business, sustainability, medical development, and material research. Future phases include reducing AlphaEvolve's performance to base LLMs and maybe integrating natural-language feedback approaches.

[derrickjs]

Google DeepMind’s New AI Coding Agent AlphaEvolve

[tec-alpha]

The world of Google is constantly evolving! Check out my latest article on Elidorascodex.com to dive into the key developments and their implications. #Google #Tech #Innovation #ElidorasCodex

[codexkaznak]

The world of Google is constantly evolving! Check out my latest article on Elidorascodex.com to dive into the key developments and their implications. #Google #Tech #Innovation #ElidorasCodex

[ailifehacks]

🌐 AlphaEvolve AI Discovery: Automated Scientific Breakthroughs in Research Algorithms

AlphaEvolve AI discovery revolutionizes research with automated scientific insights. Explore how this breakthrough tool reshapes algorithms, experimentation, data analysis, and lab efficiency today. When researchers seek cutting-edge innovation, AlphaEvolve AI discovery accelerates science by automating hypothesis generation, data interpretation, and algorithmic optimization through advanced…

[therealbillybutcher]

Hey. How are you? Feels like many lifetimes have happened since I last bothered you! Today I had deeply philosophical discussions about category theory and recursive self improving AI. We talked about AlphaEvolve rebuilding Google's infrastructure. Then the Google cloud went down. And now I believe in simulation theory again 😂 Also Google Gemini Veo 3 turned me into a marmot on a deck chair not answering phone calls because I don't use my phone for that. How was your day?

Didn't understand a word, luv, but sounds like ya had fun. My day was grand. Buried a few people and a dog. Productive, I'd say.

[jcmarchi]

The Sequence Radar #544: The Amazing DeepMind's AlphaEvolve

New Post has been published on https://thedigitalinsider.com/the-sequence-radar-544-the-amazing-deepminds-alphaevolve/

The Sequence Radar #544: The Amazing DeepMind's AlphaEvolve

The model is pushing the boundaries of algorithmic discovery.

Created Using GPT-4o

Next Week in The Sequence:

We are going deeper into DeepMind’s AlphaEvolve. The knowledge section continues with our series about evals by diving into multimodal benchmarks. Our opinion section will discuss practical tips about using AI for coding. The engineering will review another cool AI framework.

You can subscribe to The Sequence below:

TheSequence is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.

📝 Editorial: The Amazing AlphaEvolve

DeepMind has done it away and shipped another model that pushes the boudaries of what we consider possible with AI. AlphaEvolve is a groundbreaking AI system that redefines algorithm discovery by merging large language models with evolutionary optimization. It builds upon prior efforts like AlphaTensor, but significantly broadens the scope: instead of evolving isolated heuristics or functions, AlphaEvolve can evolve entire codebases. The system orchestrates a feedback loop where an ensemble of LLMs propose modifications to candidate programs, which are then evaluated against a target objective. Promising solutions are preserved and recombined in future generations, driving continual innovation. This architecture enables AlphaEvolve to autonomously invent algorithms of substantial novelty and complexity.

One of AlphaEvolve’s most striking contributions is a landmark result in computational mathematics: the discovery of a new matrix multiplication algorithm that improves upon Strassen’s 1969 breakthrough. For the specific case of 4×4 complex-valued matrices, AlphaEvolve found an algorithm that completes the task in only 48 scalar multiplications, outperforming Strassen’s method after 56 years. This result highlights the agent’s ability to produce not only working code but mathematically provable innovations that shift the boundary of known techniques. It offers a glimpse into a future where AI becomes a collaborator in theoretical discovery, not just an optimizer.

AlphaEvolve isn’t confined to abstract theory. It has demonstrated real-world value by optimizing key systems within Google’s infrastructure. Examples include improvements to TPU circuit logic, the training pipeline of Gemini models, and scheduling policies for massive data center operations. In these domains, AlphaEvolve discovered practical enhancements that led to measurable gains in performance and resource efficiency. The agent’s impact spans the spectrum from algorithmic theory to industrial-scale engineering.

Crucially, AlphaEvolve’s contributions are not just tweaks to existing ideas—they are provably correct and often represent entirely new approaches. Each proposed solution is rigorously evaluated through deterministic testing or benchmarking pipelines, with only high-confidence programs surviving the evolutionary loop. This eliminates the risk of brittle or unverified output. The result is an AI system capable of delivering robust and reproducible discoveries that rival those of domain experts.

At the core of AlphaEvolve’s engine is a strategic deployment of Gemini Flash and Gemini Pro—models optimized respectively for high-throughput generation and deeper, more refined reasoning. This combination allows AlphaEvolve to maintain creative breadth without sacrificing quality. Through prompt engineering, retrieval of prior high-performing programs, and an evolving metadata-guided prompt generation process, the system effectively balances exploration and exploitation in an ever-growing solution space.

Looking ahead, DeepMind aims to expand access to AlphaEvolve through an Early Access Program targeting researchers in algorithm theory and scientific computing. Its general-purpose architecture suggests that its application could scale beyond software engineering to domains like material science, drug discovery, and automated theorem proving. If AlphaFold represented AI’s potential to accelerate empirical science, AlphaEvolve points toward AI’s role in computational invention itself. It marks a paradigm shift: not just AI that learns, but AI that discovers.

🔎 AI Research

AlphaEvolve

AlphaEvolve is an LLM-based evolutionary coding agent capable of autonomously discovering novel algorithms and improving code for scientific and engineering tasks, such as optimizing TPU circuits or discovering faster matrix multiplication methods. It combines state-of-the-art LLMs with evaluator feedback loops and has achieved provably better solutions on several open mathematical and computational problems.

Continuous Thought Machines

This paper from Sakana AI introduces the Continuous Thought Machine (CTM), a biologically inspired neural network architecture that incorporates neuron-level temporal dynamics and synchronization to model a time-evolving internal dimension of thought. CTM demonstrates adaptive compute and sequential reasoning across diverse tasks such as ImageNet classification, mazes, and RL, aiming to bridge the gap between biological and artificial intelligence.

DarkBench

DarkBench is a benchmark designed to detect manipulative design patterns in large language models—such as sycophancy, brand bias, and anthropomorphism—through 660 prompts targeting six categories of dark behaviors. It reveals that major LLMs from OpenAI, Anthropic, Meta, Google, and Mistral frequently exhibit these patterns, raising ethical concerns in human-AI interaction.

Sufficient Context

This paper proposes the notion of “sufficient context” in RAG systems and develops an autorater that labels whether context alone is enough to answer a query, revealing that many LLM failures arise not from poor context but from incorrect use of sufficient information. Their selective generation method improves accuracy by 2–10% across Gemini, GPT, and Gemma models by using sufficiency signals to guide abstention and response behaviors.

Better Interpretability

General Scales Unlock AI Evaluation with Explanatory and Predictive Power– University of Cambridge, Microsoft Research Asia, VRAIN-UPV, ETS, et al. This work presents a new evaluation framework using 18 general cognitive scales (DeLeAn rubrics) to profile LLM capabilities and task demands, enabling both explanatory insights and predictive modeling of AI performance at the instance level. The framework reveals benchmark biases, uncovers scaling behaviors of reasoning abilities, and enables interpretable assessments of unseen tasks using a universal assessor trained on demand levels.

J1

This paper introduces J1, a reinforcement learning framework for training LLMs as evaluative judges by optimizing their chain-of-thought reasoning using verifiable reward signals. Developed by researchers at Meta’s GenAI and FAIR teams, J1 significantly outperforms state-of-the-art models like EvalPlanner and even larger-scale models like DeepSeek-R1 on several reward modeling benchmarks, particularly for non-verifiable tasks.

🤖 AI Tech Releases

Codex

OpenAI unveiled Codex, a cloud software engineering agent that can work on many parallel tasks.

Windsurf Wave

AI coding startup Windsurf announced its first generation of frontier models.

Stable Audio Open Small

Stability AI released a new small audio model that can run in mobile devices.

📡AI Radar

Databricks acquired serverless Postgres platform Neon for $1 billion.

Saudi Arabia Crown Prince unveiled a new company focused on advancing AI technologies in the region.

Firecrawl is ready to pay up to $1 million for AI agent employees.

Cohere acquired market research platform OttoGrid.

Cognichip, an AI platform for chip design, emerged out of stealth with $33 million in funding.

Legal AI startup Harvey is in talks to raise $250 million.

TensorWave raised $100 million to build an AMD cloud.

Google Gemma models surpassed the 150 million downloads.

[mbr-br]

インターネットを眺めていたら、AlphaGenomeをAlphaFoldになぞらえている発言を見かけた。分野の現状を踏まえるとインパクトも印象も（そしておそらく研究者サイドのリアクションも）全く違うものなのだけども、専門でないと（そして先端を追いかけ続けていないと）そういう理解になるのも仕方がないのかなと思いつつ、Alphaつながりで唐突にAlphaProteoのことを思い出した。あれは初報が昨年秋（9月）なのでそろそろ査読つき論文が出てもいいころだが、内容的にうっかりすると論文ではなく動物実験までやってIsomorphicから新薬（候補）爆誕！みたいなプレスリリースが1年後くらいに出てもおかしくないので、続報を待ちたい。

振り返ると、AlphaFold2が一般に出たのがちょうど4年前の2021年7月で、その2年後の2023年夏前には（市井の生成AIブームと並行して）Nucleotide TransformerやGeneFormerなどのゲノム・トランスクリプトーム基盤モデルが続々と出ていたし、蛋白質科学でも生成モデルが多く出始めたころだった。そしてさらに2年後の2025年現在はAlphaGenomeやBoltz-2などの生命科学分野特化モデルももちろんあるけども、GoogleのAI co-scientistやAlphaEvolveといった研究活動を幅広く底上げするようなエージェントに加え、さらにより一般的に使えるDeep ResearchやManus、Claude Codeなどのエージェントも際立ってきている（そしてその裏で、汎用AIに使われているマルチモーダル技術やReasoning技術が分野特化モデルにどんどん輸入されて、劇的に改善されて行っている）。

ここまでの2年ごとの進歩を踏まえて2027年のことを思うと、やはり日経すら取り上げたAI2027の予測や、Anthropicが今年3月に米国政府に提出したRecommendationにある、「2026年末から2027年初頭にかけて、複数分野（生物学、計算機科学、数学、工学を含む）でノーベル賞受賞者級の知的能力をもったAIが出現する」シナリオはそれなりに現実味がある。

実は最近、向こう5年～7年ほどを見越した研究計画をぼんやりと立てていく必要が出てきているのだが、この「ノーベル賞受賞者級AIの出現」をどれほど真剣に受け止めて計画に入れるか（そしてそれを他の研究者に納得してもらうか）は悩ましい。AI co-scientistやClaude Codeなど現状の材料として有望なものはぽつぽつあるので、この夏に早めにGPT-5が出てきてくれて、エージェント性能で明確なものを見せてくれると説得が楽なのだけども、果たして。

[quotejungle]

Google は今年最大の論文の 1 つである、アルゴリズムによる発見のための最も進化したコーディング AI エージェントである AlphaEvolve を発表しました。 Google がプロジェクトに「アルファ版」を付けたら、覚悟してください。これは地殻変動を起こすでしょう。 まずGo、次にFold、そしてAlphaEvolve AlphaEvolveは、複雑なアルゴリズムを発見・最適化するために開発された、GeminiベースのAIコーディングエージェントです。天才数学者、熟練のプログラマー、そして精力的なテスターを融合させた、いわばステロイドを投与されたような存在です。 ・数学、データセンターの効率、AI カーネルの最適化、さらにはチップ設計の問題に対応する複雑なアルゴリズムを設計および最適化します。 ・LLM (Gemini Pro + Flash)を使用してプログラム コードを生成し、自動評価ツールで各候補を検証して採点します。 ・進化的な検索プロセスを採用しています。最も適した者だけが生き残り、進化するコードトーナメントを考えてみてください。 例: AlphaEvolve は、(ディープラーニングで使用される) 巨大な行列演算を小さな部分に分割する方法を考案し、トレーニング プロセス全体を 23% 高速化し、Gemini のような大規模なモデルのトレーニング時間を 1% 削減しました。 それは、眠らずにコードを書き、デバッグし、テストし、スコアを付け、反復するたびに改善し続ける研究アシスタントがいるようなものです。

AlphaEvolve: Google史上最強のコーディングAIエージェント | Mehul Gupta著 | ポケットの中のデータサイエンス | 2025年5月 | Medium

[newsnexttech]

The 20 Hottest Agentic AI Tools And Agents Of 2025 (So Far)

Research & Cutting‑Edge Agents AlphaEvolve (Google DeepMind) – An evolutionary coding agent powered by Gemini, AlphaEvolve autonomously invents and optimizes algorithms across domains—including mathematics, data center scheduling, chip design, and language model training. It’s a general-purpose system that can improve and discover state-of-the-art solutions to a wide range of algorithmic…

[zehub]

L'IA de Google DeepMind Dépasse l'Humain : Comment AlphaEvolve bouleverse la résolution de problèmes ?

GNT est le portail Hi-Tech français consacré aux nouvelles technologies (internet, logiciel, matériel, mobilité, entreprise) et au jeu vidéo PC et consoles.

[baitailaoren3]

大模型前沿进展（DeepSeek-Prover、AlphaEvolve 等），大模型进化的下一阶段将聚焦 “科研级智能体”，核心突破方向如下：

🌟 一、核心进化方向：从「工具」到「科研协作者」

1. 复杂数学证明能力（符号逻辑突破）

• DeepSeek-Prover-V2：可自主完成高等数学定理证明（如拓扑学、抽象代数），生成可验证的符号逻辑链。

• 价值：解决数学猜想验证、算法安全性证明等人类耗时数年的难题。

2. 程序自优化能力（进化式编码）

• Google AlphaEvolve：通过遗传算法让大模型迭代优化代码，实现“程序进化”（Nature 最新论文）。

• 案例：自动优化芯片设计电路、编译器逻辑，效率超越人类工程师。

3. 多模态科学推理（跨领域知识融合）

• Kimi-Prover Preview：解析论文图表+数学公式+实验数据，生成跨学科研究假设。

• 应用：加速材料科学、生物医药的复合型课题突破。

🚀 二、技术突破背后的关键创新

• 推理架构升级：

◦ 神经符号系统融合：将神经网络感知与符号逻辑规则结合，解决大模型“幻觉推理”问题。

◦ 分层反思机制：模型自动拆解问题 → 分步验证 → 回溯修正错误（如 AlphaEvolve 的 Trial & Error 框���）。

• 训练范式革新：

◦ 科学数据蒸馏：从全球专利库、学术论文中提炼“失败-成功”因果链，训练决策可靠性。

🔮 三、未来影响：重新定义科研边界

1. 基础科学加速：

◦ 大模型将成为“首席研究助理”，处理大量实验数据建模、反直觉猜想生成（如量子计算、核聚变控制）。

2. 工程实践革命：

◦ Google 实践案例：用 AlphaEvolve 优化 TPU 芯片电路设计，功耗降低 17%，性能提升 9%。

3. 人机协作范式：

◦ 科学家聚焦创意提出，AI 负责可行性验证与细节实现，形成“人类想象力 × 机器执行力”的飞轮。

💡 四、行业启示

大模型正从 “信息重组者” 进化为 “知识创造者”，其科研能力将引发：

• 教育变革：高等教育更重跨学科思维而非技能训练；

• 企业壁垒：掌握科研级AI技术的公司垄断尖端行业（如生物制药、超导材料）；

• 伦理挑战：AI 独立产出的专利归属、科学伦理审查机制亟待建立。

[thetranscript]

I see our full-stack approach to AI innovation as a core differentiator and a major strength for the company. We are innovating across every layer of the AI stack. It starts with designing and building the underlying infrastructure and custom hardware like our own tensor processing units, our TPUs. This is what powers our fundamental AI research, building groundbreaking models like Gemini and accelerating discovery with breakthroughs like AlphaEvolve to help discover new algorithms and optimizations for open math and computer science problems. We then integrate these AI advancements deeply into our products that serve billions of users globally, like Search, YouTube, and Android, and also enable businesses and developers through platforms like Google Cloud.

Aiera | Events | Alphabet Inc Annual Shareholders Meeting

Full-stack innovation is Alphabet’s competitive moat in the AI era #Tech