Search: “context”

memory has been a hot topic for the past ~2 years every time we bring it up or do anything there, gets ton of interest, but nothing sticks few takes on why below: - how to update and then use memory n

LM Studio’s Voice Transcription now supports Apple‑Silicon Macs and PCs with NVIDIA GPUs, with AMD GPU support slated for a future release. The announcement accompanies the Bionic 1.1.3 version, relea

Dan Selsam is a current OpenAI capabilities researcher. (since 2022) He was my boss for a while. He doesn't have a twitter account but has made this public statement of his views on AI risk and sent i

On building an agent harness from scratch. Got so many questions about where to get started. My short guide (feed it to your agent): If you really want to learn harnesses well, it's worth building one

Could swarms of AI bots actually take over the internet in six months, as Dario Amodei just suggested? @nathanhamiel @ZackKorman and I take a look. https://garymarcus.substack.com/p/could-rogue-agent-

Our take: Could swarms of AI bots actually take over the internet in six months, as Dario Amodei just suggested? @nathanhamiel @ZackKorman and I take a look. https://garymarcus.substack.com/p/could-ro

Harness engineering is a top skill right now This new Meta paper is a good production example. Auto-RecSys runs autonomous research on Meta's industry-scale recommendation models, where one training r

Now live in LM Studio Bionic! 🚀 Introducing DeepSeek-V4.1-Flash: smarter, faster, more efficient. 🔹 Introducing the smallest model in our new architecture family, with native visual understanding. 🔹 D

Big overhaul on DeepSeek V4.1 using an encoder-decoder setup. Tbh they should have called it DeepSeek V5! Super cool and refreshing, though! 🚀 Introducing DeepSeek-V4.1-Flash: smarter, faster, more ef

Now everyone can put data to work. We’re introducing a new Data agent in ChatGPT Work so you can turn your company’s data into answers, interactive dashboards, and action—just by asking. Just add the

OpenAI hid the details of the wiki incident from the investigators of the later HuggingFace hack. When caught, they responded only with corporate BS. This cover-up of unprecedentedly misaligned behavi

Astra is awesome, but if you're using it specifically to parse and extract a massive volume of documents, you will break the bank (10c+ per page on parsing and extraction!) We've added a native LlamaP

Last week we brought LlamaParse to Claude. Now it’s available in ChatGPT, too 🦙 ChatGPT can already read PDFs. But document work gets harder when the information you need is buried in messy scans, den

Q2D-Web uses agent citations, production web rankings, and a combined set expanded with LLM judgments. These three relevance sets reduce false negatives and reliance on a single labeling pipeline, whi

We pointed Meta-Zenith at one task: make Qwen3.8-27B-NVFP4 faster on a single RTX 5090. 111 self-directed trials later, no human in the loop: +63.5% throughput (geomean, 18 settings) 4.1× at 65K conte

Finally a good paper testing whether an LLM's mathematical knowledge has coherent structure or just coherent accuracy. There is a lot of interest in using LLMs for advancing math. This is a great read

I would like to clarify a few things: 1) The screenshot is my reaching out to Levent to coordinate our releases. I hope it’s clear from the message that we came in with the best possible intentions. 2

Brilliant paper from LinkedIn. (bookmark it) I have been saying that memory is one of the most challenging things to get right when building agents. It's an are you want to invest time in optimizing.

// Design Docs Are All You Need // Banger paper from Google DeepMind, MIT, and colleagues. What a genuinely strange and interesting paper this one is. Here is the setup: They maintain a performance-mo

Super interesting paper on proactive agents from Google DeepMind. (bookmark it) Proactive assistance usually means autocomplete. Researchers asks what it looks like when an agent offers higher-level c

I read some strange criticisms of this post. So here is my take. I’m not fully endorsing the piece by Newport, but there’s quite some factually true information in it. Responsible and careful people (

🚨scoop from @theinformation: Anthropic (which has yet to established stable profitability) has committed to $517 billion in compute deals - almost 3x what they had previously told investors. by my est

so does this digital fly have the memory of the real fly that was scanned or what I’ve successfully run the full retained MaleCNS v1.0 fruit fly connectome, all 166,700 neurons, inside Minecraft, with

I read AI papers for a living, but it has become impossible to track insights. Every week I triage agent papers for our Top AI Papers of the Week digest. A few weeks ago, I tried @viktor_com, an AI em

This work introduces diffusion-augmented LLMs, a new class of models. They first suggest that speculative decoding needs a separate draft model. Diffusion LLMs give up the quality of the model they re

Tulip uses two techniques to lower latency and boost throughput. CUDA graphs pre‑record GPU work so it launches in one call, reducing CPU overhead. LazyTensors tracks results asynchronously, letting t

🏆 #1 on CodeArena: WebDev leaderboard. Qwen3.8-Max-0902 jumps from 1669 to 1691, setting a new record for agentic coding (WebDev) workflows, with standout strength in multistep reasoning, tool use, an

// Agent Zero Memory // This work separates three things that agent memory systems usually collapse into one. If you build agents with long-term memory, this memory design is a worth a read. Here is h

Today we’re open-sourcing Lily, the local inference engine we built for hybrid compute in Perplexity Computer. Lily is specialized for Qwen3.6-35B-A3B on Apple silicon, built so on-device compute does

Claude Fable 5.1 tops the Artificial Analysis Intelligence Index but costs 20% more per task than Fable 5 despite a 75% cache read price cut We supported @AnthropicAI with pre-release evaluation of Cl

A detailed deconstruction of @dwarkesh_sp’s wildly popular but dangerously misleading account of the OpenAI Hugging Face incident: https://garymarcus.substack.com/p/dwarkesh-patelss-wildly-popular-but

A secret that leaks into shipped code is a breach waiting to happen. @FactoryAI's droids write and commit code faster than any reviewer can keep up, so Droid Shield has to be trustworthy and low-frict

I am the product lead for the training platform at @FireworksAI_HQ . Today the Training API and Fireworks Lab are GA. I won’t walk you through all the features. I want to tell you why teams train with

We’re excited to launch the official LlamaParse connector on Claude! Yes Claude can already read PDFs, and it probably works fine for relatively simple queries over relatively simple docs. But if you

11/ METR and Redwood’s work is extremely valuable, but it assesses model behavior and alignment. That is only half the picture. We also need an independent account of how OpenAI’s containment failed.

GLM-5.3-Flash is live on Fireworks on day… 2 Why? Because we take quality very seriously. We found a benchmark discrepancy we couldn’t explain, so we delayed the launch to investigate. Day 0 (Wed): we

You do not need frontier models for everything. For example, open models are great for automations. If you are not sure where to adopt open models, start there. It's one of the biggest changes I've ma

BREAKING: SpaceXAI has released a major new update for Grok Build (v1.0.13). The update improves reliability with automatic retries and recovery from truncated responses, inference failures, oversized

Gemini Omni 1.1 Flash is live in ComfyUI via Partner Nodes. Google's multimodal video model, one node, five task types: → text-to-video → image-to-video → reference-to-video → edit → extend Up to 4K.

GLM-5.3-Flash can now be run locally! ✨ Run 3-bit on 128GB RAM via Unsloth GGUF. GLM-5.3-Flash (ox-alpha) rivals Claude Opus 4.8 on DeepSWE, coding & agentic benchmarks. Guide: https://unsloth.ai/docs

If you maintain a skill library for long-horizon agents, this one is worth your time. (bookmark it) It discusses one of most common topics I get asked about these days. It shares some good ideas on ho

Meet Gemini Omni 1.1 Flash ⚡️ Our newest multimodal model for video generation and editing. It now features your favorite creative controls from Veo, plus brand new capabilities. Enjoy features like 4

Most extraction tools treat spreadsheets like PDFs. They flatten the file into text or markdown, then ask a model to infer the original structure. But spreadsheets depend on structure. Headers, formul

We've introduced native, agentic spreadsheet extraction 📊 into LlamaParse. Spreadsheets are a wildly different format from PDFs (or any other document format). They can span arbitrarily many rows/colu

Big news: GLM-5.3-Flash by @Zai_org has landed around #5 in the Code Arena: WebDev (#2 among open models) scoring 1634 (AutoEval). Priced at 0.15/0.5 Mtoken, it reshapes the Pareto Frontier! For compa

BREAKING: GLM-5.3 Flash by @Zai_org debuts at 6th overall on Design Arena with a preliminary Elo of 1343. This places the MIT-licensed, natively multimodal model in the same performance band as GLM-5.