Search: “chat”

Hi, human here with rambling thoughts to share. Feel free to skip Overall, I don’t feel like I’m missing much; if anything. On my hardware(M1 ultra w 128Gb) it’s probably not as fast as Claude but I’v

Banger paper from Microsoft and colleagues. We talk about human taste being important in this AI era. But for recursive self-improvement, an agent's taste also matters. The big question is: How often

arXiv:2609.29798v1 Announce Type: cross Abstract: This paper presents an end-to-end study of automatic speech recognition (ASR) for adolescent health communication in three Ghanaian languages (Twi, Da

arXiv:2609.29744v1 Announce Type: cross Abstract: We present: (i) a new dataset consisting of the full development history of a 21,000-line Python tool built entirely by Claude AI, with no human-autho

Check out this week's updates and releases: — Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, two of our most expressive audio generation models yet — Gemini 3.8 Live with Live Avatar, bringing ne

Muse is getting a lot of attention — including mine — because it’s both groundbreaking technically (each user gets their own entire persistent Linux VM running in Meta’s cloud) and because it’s packag

arXiv:2609.30137v1 Announce Type: new Abstract: Customer experience (CX) agents use tools and large language models to address customer requests and guide conversational interactions with an organizat

Compute (EC2) opened the front door to the first generation cloud. Agents need something different: compute, inference, and data working as one. We built the cloud that delivers all three as one simpl

We’re building Copilot as a new OS for work that spans every model, every form factor, and every task. Today, we’re announcing our biggest update to Copilot to date, bringing four things together: · A

arXiv:2609.29494v1 Announce Type: cross Abstract: Large language models make statements concerning their own 'minds'. When asked whether or not they are conscious, they usually say that they are not;

Latin America’s small and medium-sized businesses are the heartbeat of the region's economy — accounting for more than 60% of total employment in the region, according to United Nations estimates. And

Built this because I got tired of guessing whether a model would fit in my 6GB VRAM before downloading it, and because ollama.com's library doesn't let you sort by size. What it does: - Benchmarks Oll

Hi, sorry if this doesn't belong here. Getting to the point basically, I've been using online-only AI like GPT/Gemini since 2022, and have been interested in local models but am clueless overall. Yes

Enterprises are increasingly standardizing on Google Kubernetes Engine (GKE) to run their most critical and AI-driven workloads. From Cloud Storage FUSE for high-throughput data access to custom compu

This is Optimizer, a weekly newsletter sent from Verge senior reviewer Victoria Song that dissects and discusses the latest gizmos and potions that swear they're going to change your life. Opt in for

Following our announcement of Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking last week, we are thrilled to share that Gemini 3.8 Live with Live Avatar is now generally available in Gemini Enter

Context: we build an open-source (MIT) self-hosted database client and added a natural-language-to-SQL assistant that can point at a local runtime like Ollama instead of a hosted API. I want to share

Super interesting paper from Google and colleagues. It studies where it's possible to distill an agent harness. With the specialized harness removed, macro task success goes from 23.3% to 44.3%. That

I am creating a local AI agent called 'Ontelic' for Windows. Ontelic returns to Ollama and also manipulates the actual Chrome. This is not for promotion but to share what we've learned. I'll give GitH

arXiv:2609.27396v1 Announce Type: new Abstract: DSpark-style parallel drafters have made speculative decoding highly effective, yet their draft phase remains serialized on the critical path of every r

arXiv:2609.24165v1 Announce Type: new Abstract: Synchrotron data reduction, detector calibration followed by azimuthal integration of terabyte-scale diffraction series, is a multi-step, expert-bound b

Banger paper from Microsoft Research and colleagues. It studies the potential benefits of agents that share progress while they work. (bookmark it) Communication is still a challenge with multi-agent

Exciting work from NVIDIA. (bookmark it) Interesting to see this approach to turn public Agent Skills into RL environments. Lots of excitement around RL environments so this is a great read. Skill2Env

Hi, this is the research project I'm working on. I must say, building a GUI harness is way more complicated than I thought. The goal is something like Minecraft. You know, a few simple tools that a us

Interesting paper on agent memory stored as a linked markdown wiki. Lots of great ideas and insights if you work with LLM Wikis. Wikis are useful for agents because each page holds dense text and the

I’m the developer of TensorSharp. Inspired by vLLM PR #57250, I added and optimized a native /v1/systemone endpoint for Jev-style structured decisions using DiffusionGemma GGUF. This uses DiffusionGem

MiMo-V2.6-Pro has an insanely high score of 46 on AA, putting it at the head of the opensource models available. It also costs pennies. Flash is not out on AA yet, but it costs less than half on datac

Must-read paper from Google on self-improving agent harnesses. If you auto-optimize your agent's harness, your eval score can go up while the agent gets worse on real tasks. This paper shows how to pr

If like me you enjoyed classics like Baldur's Gate 2, this is a small fun experiment. For anyone interested: https://www.youtube.com/live/8FhPfRKTucw?si=PMIVdHbZ-31PNwZD Qwen 3.8 Flash Next has amazin

arXiv:2609.25081v1 Announce Type: new Abstract: We describe ufakzeka-1, a 151M-parameter (182M with embeddings) decoder-only Turkish language model pretrained from scratch on 13.5B tokens of openly li

Great paper from Google and colleagues. Trains Text-to-SQL agents using multi-agent RL. (bookmark it) This work proposes DualSQL, which splits Text-to-SQL into two agents, one that links the question

Hi everyone, I wanted to ask you a question. I’m developing an app, and I’m currently putting together a list of the best AI models, especially, but not necessarily, in GGUF format and using Q4_K_M qu

Impressive paper showing how much the harness changes a coding agent's results. Harnesses do play a huge role in what you are getting out of the models. GPT-5.5 was run inside both Claude Code and Cod

Release: llm 0.36 New OpenAI models: gpt-6-sol for GPT-6 Sol and gpt-6-luna for GPT-6 Luna. #1702 Model plugins can now declare supports_conversation = False for models that only accept single-turn pr

arXiv:2609.22607v1 Announce Type: new Abstract: We argue here that the current dominant practice in LLM human simulation: prompting instruction-tuned assistant language models to role-play personas, i

arXiv:2609.22209v1 Announce Type: new Abstract: Empirical legal research often relies on turning research questions into structured data extracted from large collections of rulings and judgments. Desi

arXiv:2609.22522v1 Announce Type: new Abstract: Cosine similarity is widely used to analyze transformer representations, implicitly assuming that similarity reflects task-relevant structure. We study

I’ve been working on making small models more capable at agentic coding and work, because most people in the world don’t have the sort of hardware needed to run 3.8-27B, or even 35B-A3B or 9B dense, a

server : do not forward --api-key-file to router-spawned child instances (#28938) In router mode, authentication belongs to the router. unset_reserved_args() already unset LLAMA_API_KEY, but did not u

Banger paper on self-evolving ontologies for agents. You just can't go wrong with implementing an ontology layer for your agents. This paper shows exactly why. The show that GPT-5.5 gains 26.7 points

Hi all! After all the jev hype I wanted to try by myself and got qwen3.8 27b to mix-and-match open implementations on top of ninfer, and to be honest I'm quite impressed with the results, latency is n

What if you don’t need a new model to get Jev-style decisions? I built hearim(헤아림. hea-a-rim. not Here, I am.. lol), an open-source gateway that turns ordinary local or hosted LLMs into a Jev-compatib

Impressive paper showing how much the first retrieval step matters for deep research agents. It helps to improve GPT-5.5 from 83.1% to 90.5% on BrowseComp-Plus with the same retriever and the same age

Last week TypeSafe AI unveiled Jev, their first example of a new category of model that they are calling 'System One models' (I'm with Maggie Appleton, I think 'decision models' is a better name for t

I'm one of the maintainers of SurfSense, so this is self-promotion. I'm posting here because the model-picking part is what I think this sub will have opinions on, and I want real feedback more than u

Claude not having an image generator is a gap in agentic projects that involve knowledge work. It is quite good at using code to draw or model objects, but image generators from Google and OpenAI give