Search: “llm”

Grok is the Flow LLM. Grok 4.5’s biggest advantage is its speed. It’s smart enough to be comparable to the other models on most things. But that speed allows you to make little tweaks to your system s

🆕 In Code They Act, In Proof We Trust — Erik Meijer last year, @solomonstre defined agents as 'an LLM that's wrecking its environment in a loop', and @simonw coined the Lethal Trifecta for agents, tha

Hardmaru posted a tweet on July 13, 2026 stating that while language models and coding agents are valuable, “there is more to life, and more to AI, than just LLM agents.” The tweet received 61.1 k vie

1/ I watched LangChain’s webinar on LLM wikis and agent memory and wrote up a working paper from my own small agent-fleet experiments: Soft-Cache: A Human-Supervisable Coherence Protocol for Persisten

Thank you @hwchase17 @BraceSproul @devstein64 @jeffreyhuber for the LLM Wikis session today, I absolutely loved it! Really grateful you keep these sessions open and honest about what actually works in

Highly-recommended read. Aligns with what I see in my own harness: > Pi harness got the same success rate as harnesses from the LLM vendors with Opus and GPT, but at 2x less cost > GLM 5.2 was a major

We're releasing ZML/LLMD, our homegrown LLM server built on top of our homegrown high performance heterogeneous inference stack. It ships with 5 architectures out of the box: NVIDIA, AMD, Metal, Intel

llm wikis are a glimpse of the future of what agent memory looks like this blog i wrote resonated with a lot of folks will be discussing this (as well as some updates ive made to my beliefs since this

Fable 5 is an absolute beast at threejs. It's the best LLM at generating 3D simulations/worlds. Watch how I combine it with gpt-realtime-2 to generate interactive educational 3D worlds @dair_ai. Comme

The Log is the Agent 它的核心观点是：不要把日志当成 agent 的副产品，而要把 append-only event log 当成 agent 本身。 在 ActiveGraph 里，目标、规则、工具调用、LLM 响应、产物和关系都写入同一条事件日志；当前状态只是日志 replay 后投影出来的一张图。 这样带来三个关键能力： 可确定性重放 可在任意事件处分叉实验 从最终输出

Here's a cool piece of LLM lore: the original scaling laws were wrong due to a bug, which probably led to a lot of wasted compute on oversized undertrained models 🫣 (and that was before we even starte

I've been blogging a few of my Fable experiments today, most recently I had it more-or-less one-shot a CLI coding agent on top of my LLM Python library https://simonwillison.net/2026/Jul/2/llm-coding-

The most important weird thing about LLMs is that they are so general. A bigger LLM that is better at coding is also better at ideation & ethical advice & medicine & math. This isn’t true of everythin

Turn any LangGraph agent into a voice agent, no rebuilding required. A how-to on: ✅ Making LangGraph the LLM layer of a voice agent built with @pipecat_ai ✅ Tracing your entire voice agents – from aud

New paper on giving LLM agents experience that improves the weights and stays readable at the same time. Agent-experience methods split into two camps. Externalized natural-language rules stay interpr

Great report on LLM agent communication protocols. Communication is a huge bottleneck in multi-agent systems. (worth bookmarking) The report builds a five-dimensional taxonomy (counterparty, payload,

How does it work? Sakana Fugu is itself an LLM, trained to call various LLMs in an agent pool, including instances of itself recursively. Fugu dynamically orchestrates the world's best models to tackl

The largest LLM-as-a-Judge reliability audit yet. Researchers ran 21 judges from nine providers over roughly 541,000 judgments on MT-Bench, JudgeBench, and RewardBench. Findings: Validating a judge wi

'When an LLM outputs a step-by-step plan, it creates a powerful illusion that you are watching a machine reason its way to a conclusion. A position paper by professor Subbarao Kambhampati and research

Called the LLM price wars – which are about to heat up — over two years ago. See especially predictions 3, 4, and 7, which laid out the no moat, small profit, price war regime we have seen ever since.

Don't miss the exact text though: 'We’re changing Fable 5’s safeguards for frontier LLM development to make them visible' - make them visible means they're undoing the truly egregious (dare I say 'una

100%, hallucinations are baked in (as I have been saying since 2001) and that is why basically no LLM company afford to operate in Germany now. We need a better technology. @GaryMarcus Lol, and Google

first in a series of technical blogs of how we build llm infra How do you support full-text search JSON filtering over agent traces that span up to hundreds of MBs, while keeping a median (P50) latenc

I believe what Anthropic is doing, gating the ability to do certain harmless things like LLM research, and with incredibly sensitive filters that even medical questions are often blocked, is *deeply*

If the new German court decision — to hold LLM companies liable for the false claims their LLMs made — become universal, effectively sidelining generative AI until better techniques came along, would

What if, Germany locked itself out of the LLM race and • Had students who actually learned things in high school, instead of turning in prompt outputs they barely read • Emerged from the sea of slop •

mythos will be bad ON PURPOSE on ai 'frontier llm research' tasks, this is very very sad for the research community also the fact that this is un purpose not visible to the user is crazy Introducing C

Hot take: Universities charge $300,000 for a degree that teaches you skills any LLM can do for free. At some point we need to have an honest conversation about whether higher education is the greatest

This post highlights four recent improvements to the ecosystem of open-weight large language models designed to run efficiently on consumer hardware, covering developments that make local LLM deployme

New research from Google. Just shows the impressive results you can get from custom agent harnesses. LEAP wraps a general-purpose LLM in an agentic scaffold that grounds every step in the Lean compile

LangSmith's LLM Gateway provides real-time cost visibility and spend tracking capabilities, allowing users to monitor their language model usage costs aggregated by workspace. This feature helps elimi

Can we design legal agent verifiers that are up to 1,000x cheaper? Verifiers are LLM judges that check an agent’s work against rubric criteria: they're used both in agent benchmarking and as reward si

// Scaling Behavior of Single LLM-Driven Multi-Agent Systems // Does adding more agents actually make a multi-agent system better? It's possible that collective intelligence emerges from interaction d

🌞This is big Local AI news! A new open-source Computer-Use LLM has just launched. Holo 3.1 is H Company’s (🇫🇷) new local computer-use agent model that beats Qwen3.5-397B, Kimi-K2.5, and Sonnet 4.6! Si

slowly we're all realizing that tools should be called from code, not from within the llm api Introducing Search as Code, our new search architecture for AI agents. It writes Python that calls our sea

These are some of my LLM assisted contributions from the past month. Nothing amazing, but I'm slowly getting better at it. Atm, using Qwen3.6 27B exclusively. For hardware - switching between M2 Ultra

I did a talk on LLM Wikis and HTML artifacts recently, if you are curious to learn more on the topic: https://academy.dair.ai/events/cmovobp97000904l5h0n9a2yz Doing a second session and a few releases

ComfyUI just added @OpenRouter support. Instead of being locked into a single LLM, you can now access 20+ models directly inside Comfy. More flexibility, less friction, same workflow. Links to the wor

LangSmith LLM Gateway is a feature that enables proactive cost and privacy controls by enforcing spending limits and redacting personally identifiable information (PII) before API requests are sent to

pibot is now running fully local, using parakeet for STT, qwen3-tts for TTS, and Qwen 3.6 as the local multi-modal LLM via llama.cpp. The STT and TTS inference engines are Rust/mlx-c based. Ported fro

Every LLM API call depends on the inference engine underneath it. Tokenization, scheduling, prefill, decode, KV cache, batching, and streaming determine whether the experience is fast, scalable, and p

The MiniMax M2 series was one of the most widely used open-weight LLM series earlier this year. Now, we got a technical report with some interesting tidbits. I summarized some of them below: 1. Full a

LLM companies are likely to be like airline companies: small margins, intense competition, high expenses. @GaryMarcus Airlines taught us something important: unlimited demand does not guarantee attrac

not using LLM’s works for me Prolonged AI use may make it harder to think critically and creatively, recent research suggests. But there are ways to keep the brain fit https://www.economist.com/scienc

oh even better, since you can test the runtime without an LLM, you can post this into Claude: “look up http://activegraph.ai, install and run a small experiment I would like leveraging this, and expla

If this is true, using the best public estimates we have of LLM resource use, solving this Erdos problem took 0.6–6.3 kWh of electricity and about 3–31 liters of water. So that is less than three almo