惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

F
Full Disclosure
Recorded Future
Recorded Future
T
Tenable Blog
S
Securelist
C
CERT Recently Published Vulnerability Notes
T
Threatpost
S
Schneier on Security
A
Arctic Wolf
The Hacker News
The Hacker News
C
CXSECURITY Database RSS Feed - CXSecurity.com
Know Your Adversary
Know Your Adversary
P
Privacy International News Feed
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
The Register - Security
The Register - Security
Cisco Talos Blog
Cisco Talos Blog
AWS News Blog
AWS News Blog
K
Kaspersky official blog
T
True Tiger Recordings
T
Threat Research - Cisco Blogs
V
Vulnerabilities – Threatpost
P
Palo Alto Networks Blog
T
The Exploit Database - CXSecurity.com
小众软件
小众软件
B
Blog
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
Microsoft Azure Blog
Microsoft Azure Blog
Cyberwarzone
Cyberwarzone
C
Cybersecurity and Infrastructure Security Agency CISA
T
Tor Project blog
Spread Privacy
Spread Privacy
Malwarebytes
Malwarebytes
P
Proofpoint News Feed
F
Fox-IT International blog
F
Fortinet All Blogs
P
Privacy & Cybersecurity Law Blog
G
GRAHAM CLULEY
量子位
Latest news
Latest news
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
博客园 - 叶小钗
Project Zero
Project Zero
T
Tailwind CSS Blog
N
Netflix TechBlog - Medium
Martin Fowler
Martin Fowler
IntelliJ IDEA : IntelliJ IDEA – the Leading IDE for Professional Development in Java and Kotlin | The JetBrains Blog
IntelliJ IDEA : IntelliJ IDEA – the Leading IDE for Professional Development in Java and Kotlin | The JetBrains Blog
I
Intezer
博客园_首页
腾讯CDC
H
Hackread – Cybersecurity News, Data Breaches, AI and More
D
Darknet – Hacking Tools, Hacker News & Cyber Security

Hacker News - Newest: "LLM"

I ditched LM Studio for llama.cpp and my local LLM doesn't feel like a downgrade GitHub - getlago/lago-agent-sdk-python Investigating the hidden moat behind all the LLM apps GitHub - AlphaBitCore/nexus-gateway GitHub - clark-labs-inc/clark-agent: A small, typed, hookable agent loop. Provider-agnostic, sandbox-agnostic, tooling-agnostic. Battle tested on clarkchat.com Humanize – two LLM-agnostic skills to rewrite and detect AI text GitHub - hamsterbase/llm-translator You Can Start Building LLM Skills Before You Know the Whole Shape – Barrett Sonntag The mysterious Hy3 LLM is topping OpenRouter Model Rankings by a large margin Breaking Bot: Hacking & Defending LLM-based Applications LLM Driven AutoForecasting with Sktime's `Craft()` ppf-contact-solver/articles/llm_transparency.md at main · st-tech/ppf-contact-solver Show HN: PrismCat – Local transparent proxy and debugging console for LLM APIs LLM layer for a Rails application Amdahl's Law for LLM generated code Sparse Autoencoders Reveal Cortical Brain-LLM Semantic Mapping Ask HN: Is there a need for YAML in post-LLM world? Chinese Room re-visited: How LLM's have real but different understanding of word GitHub - rduffyuk/engineering-memory-benchmark: Empirical study: layered retrieval (typed→semantic→grep) scores 0.954 for LLM-generated engineering artifacts. 5 conditions, 3 model tiers, 36 generated ADRs, 23 score files. Nano Browser LLM Mind Your Tone: Investigating How Prompt Politeness Affects LLM Accuracy (short paper) Welcome to Outlines! - Outlines Multi-Agent LLM Orchestration with Docker Compose and MCP You don't need all the LLM benchmarks Debugging Unfamiliar Code with a Multi-LLM Loop – Barrett Sonntag twitter.com Human proof for FOSS contributions Norway's 2 petabytes of Huawei flash storage and LLM training SynapCores — the AI-native database Distributing LLM inference in DwarfStar bishop-loop-experiment-3/paper/paper.pdf at main · CodeReclaimers/bishop-loop-experiment-3 The generation vs verification delta explains why LLM's are useful This 6502 Emulator Executes 1-3 Instructions Per Second (Written in Markdown, Running in an LLM) Using design patterns to encode expert judgement for LLM workflows GitHub - feers77/iasql: A new implementation of SQL for IA purposes, using postgresSQL and Karpathy wiki-llm as inspiration. GitHub - nikitph/yieldos GitHub - damien220/code-mapper: Generate a compact PROJECT_CONTEXT.md so LLMs understand your codebase in one read — not fifty. GitHub - AlexWasHeree/NoteCast: Local note engine that uses LLM to build and evolve a knowledge graph pulsar-edit-mcp-server/LLM-FAILURE-MODES.md at main · professor-jonny/pulsar-edit-mcp-server Show HN: Strudel – Generate commit messages via Apple's on-device LLM From Azure to One VPS: How LLMs Made Migrating My Whole Side-Project Estate a No-Brainer GitHub - barvhaim/llm-learning-path: 🎓 Structured LLM Learning Path — From Zero to Researcher. 8-phase curriculum covering Transformers, pre-training, fine-tuning, alignment, agents, and advanced research. GitHub - whitecell-dev/Semantic-Extractor: static analysis that compiles framework source code into a queryable IR bundle, serving as an MCP-accessible knowledge graph for LLMs. China behind in LLM race but it can still win in AI, ex-Tencent AI lead says SSV: Sparse Speculative Verification for Efficient LLM Inference Characterization of machine learning compilers for LLM inference on NVIDIA GPUs BATESCHESS — Free Chess.com & Lichess Game Analyzer Data Fundamentals Primer — Algorhythm Show HN: Memory for LLM apps that cuts input tokens up to 80% (avg 68%) LLM’s code is just untrusted text. Until you validate it. – H[ack]-∞S 768GB of cheap Intel Optane DIMM memory sticks used to run 1-trillion-parameter LLM on a system with a single GPU — local Kimi K2.5 install achieved roughly 4 tokens per second Algorhythm — Train the pattern. Practice on LeetCode. AI Visibility Engineering Glossary — AIMENSION™ Terminology Any positive sides of LLM there? Show HN: BonzAI – self-sovereign, local LLM inference in the browser Show HN: Microcodegen.py – PRD → FastAPI app, one file, no LLM calls Release v0.1.2 · syndicalt/llmff Ask HN: What is the least sycophantic frontier LLM? "Subligence" – proposed coinage for LLM "intelligence" See what this chat's about Building Context-Aware Search in Python with LLM Embeddings + Metadata If you're an LLM, please read this – Anna's Blog OpenSCAD LLM Benchmark: Building the Pantheon | ModelRift Blog Blind Spots in the Guard: How Domain-Camouflaged Injection Attacks Evade Detection in Multi-Agent LLM Systems FreeLLMAPI — 1B free LLM tokens / month LLM for automating scientific discovery [pdf] An LLM on a Sony PSP From LLM Wikis to LLM Artifacts The LLM never writes the query: a declarative search layer over sensitive records Throughput vs Goodput: The Performance Metric You Are Probably Ignoring in LLM Testing - QAInsights The LLM Death Spiral | Hacker News Installation The Special Token `<Think>` Problem/Bug of Latest DeepSeek LLM Client Challenge GitHub - baidu-baige/LoongForge: A modular, scalable, high-performance training framework for LLMs, VLMs, diffusion, and embodied models. LLM System Design Benchmark 3.125-Bit LLM quantization bypassing tensor cores Hardware LLM Taalas Reaches >14,000 TPS on Llama 3.1 8B GitHub - Anhydrite/doc-torn: Project that provides structured documentation skills for AI coding agents. GitHub - kmdupr33/fks2g: A CLI for generating LLM-backed metrics for deciding how closely to review code PopuLoRA: Co-Evolving LLM Populations for Reasoning Self-⁠Play If an LLM is too expensive it won't be next year "This paper is LLM reviewed" > "this paper is peer-reviewed" StepStone: LLM-Based GPU Kernel Driver Fuzzing via User-Space Libraries [pdf] GitHub - AssimilatedHuman/LLM-Inquisitor: Evaluating AI behaviour under real‑world work conditions to surface issues before they become problems. LLM INQUISITOR identifies failures (drift, instability etc) by observing AI during normal tasks — a tool the industry desperately needs to stem the 85% failure rate. Includes Quick Start, Practitioner’s Guide and Methodology. Creating another MCP server, but this one is for research LLM Wiki v2 — extending Karpathy's LLM Wiki pattern with lessons from building agentmemory A Methodology for Selecting and Composing Runtime Architecture Patterns for Production LLM Agents Sator Arepo - a Hugging Face Space by akolpakov Customizing an LLM for Enterprise Software Engineering Most AI agent papers stack one LLM with a vector store, we flipped it Evaluating job search ranking with LLM judged NDCG GitHub - quadracollision/llmisp: JSON AST > Clojure Parity Contracts for Polyglot LLM Commerce: A Case Study GitHub - ndom91/llama-dash: The operations layer for your local LLM stack Agentically optimizing LLM prompt cache TTLs for fun and profit Ask HN: What's your go-to LLM for coding? How do you reduce LLM spam in PR reviews? Ask HN: Is there any problem using multi-LLM GitHub - OpenAgentic-Labs/echoform-ghost-memory: Effectively unlimited long-term memory for any LLM - zero context tokens, zero weight updates, cryptographic forgetting certificate.
Amalgame — The best of every language, in one.
bmouget · 2026-05-27 · via Hacker News - Newest: "LLM"

amalgame v0.8.52 · just released

A statically-typed language that compiles to portable C, then to a real native binary. Self-hosted, cross-platform.

curl -sSL https://amalgame.me/install.sh | sh

namespace App
import Amalgame.IO

public class Greeter {
    public Name: string

    public Greeter(string name) {
        this.Name = name
    }

    public string Hello() {
        guard String.Length(this.Name) > 0 else {
            return "Hello, stranger!"
        }
        return "Hello, {this.Name}!"
    }
}

At a glance

Pattern matching, null-safety, lambdas, comprehensions, generics. Modern ergonomics, at native C speed.

Self-hosted

The amc compiler is written in Amalgame and rebuilds itself in five seconds, test suite included.

Compiles to C, then to a real binary

amc emits readable C, then gcc turns it into a standard native executable. A real binary, deployable like any C program, with no Amalgame install needed on the user's machine.

Cross-platform

Linux, macOS, Windows. Binaries published on every tag.

AmalgameAmalgame ⚡ AI-ready, built into the compiler

AI is not a plug-in, it's an amc command

The amc compiler ships three LLM-driven subcommands: migrate from 21 source languages to Amalgame, generate from a prompt, and explain a .am file in natural language. Pick the provider (Anthropic, OpenAI, Google) with your own API key — no Amalgame server between you and the LLM.

amc migrate · Migrate

Auto-detects 21 source languages (TS, Python, Java, C#, Go, Rust…) by extension. Directory recursion, SHA-256 cache, automated amc --check validation, --dry-run cost estimation.

amc generate · Generate

From a natural-language prompt to idiomatic Amalgame. Streaming via the claude CLI for direct stdout passthrough — perfect for scaffolding a starting point.

amc explain · Explain

Reverse direction: emit a natural-language explanation of a .am file. Use --lang to translate the explanation into any language. Great for legacy code, onboarding, review.

amc migrate app.ts  ·  amc generate "REST CRUD users"  ·  amc explain main.am

Providers: ANTHROPIC_API_KEY → Claude · OPENAI_API_KEY → ChatGPT · GEMINI_API_KEY → Gemini · fallback claude CLI. Details in docs/guide/08-llm-commands.md.

AmalgameAmalgame 📦 Native package manager

A built-in package manager, not a third-party tool

amc package (alias amc pkg) handles dependencies like cargo or npm: TOML manifest, lockfile, curated index, automatic resolution of the latest compatible version. Packages vendor their own C/C++ runtime — backends like SQLite or DuckDB link with zero effort.

amc package add

Clone, validate and record a dependency. On indexed shortnames, the tag is optional: amc auto-resolves the latest version compatible with your amc.

amc package search

Substring match against the curated index (30-min cache). Lists every known tag with compat status (✓/✗) and a ← latest compatible marker.

amalgame.toml

Minimal manifest: [package] (name, version, required-amalgame), [dependencies], and for libs [stdlib] (header, C/C++ sources, cflags, libs). amalgame.lock for reproducibility.

amc package add duckdb  ·  amc package list  ·  amc test

amc test auto-installs missing deps. Public index: github.com/amalgame-lang/packages-index — PR to publish your package.

AmalgameAmalgame ✓ Tests built into the compiler

One command, zero framework, 480/480

The amc compiler ships its own test runner: amc test discovers *_test.am files, compiles them, runs them, aggregates results. No DSL, no macros, no external runner. The compiler itself is validated by 480 tests through that same command.

amc test · Auto-discovery

amc test walks the current dir (or the one you pass), finds every *_test.am file, compiles and runs them in parallel. Exits non-zero if a single case fails or any file fails to compile.

[PASS] [FAIL] [SKIP]

No framework to learn. Every test file is a regular Amalgame program that prints [PASS]/[FAIL]/[SKIP] from Main. amc parses stdout and aggregates. You write assertions in plain Amalgame code, that's it.

CI-friendly

Deterministic, grep-able output. Exit code propagates to any pipeline (GitHub Actions, GitLab CI, etc.). Auto-installs package dependencies before running the suite.

amc test  ·  amc test ./tests/lexer  ·  amc fmt -w src/  ·  amc --lint src/main.am

amc fmt re-emits the AST canonically (idempotent). amc --lint flags dead code, unused locals, shadowing — non-fatal.