Search: “models”

babe, wake up. Hugging Face just dropped a new banger article about RL training coding agents on different harnesses. 1/ Excited to release The ultimate guide to multi-harness RL The same model behave

Victor M (@victormustar) tweeted on September 22, 2026 that the MiMo V2.6 release now includes an official Distill‑Qwen‑9B variant. The tweet expresses enthusiasm for this new model iteration and remi

We benchmarked DeepSeek V4.1 Flash by @deepseek_ai . It reached 98% of GPT-6 Astra’s score at 1.4% of the cost on everyday design tasks based on user requests. Every model except Astra scored lower AN

New open-source agent harness just landed! I got early access to TrueForge by TrueFoundry and have been running it locally for the past few days. The harness layer deserves as much attention as the mo

Qwen3.8-2.4T-A95B is live on Together AI. The Qwen team’s new open-weight MoE packs 2.4T total parameters with 95B active, a 1M context window, and strong coding & agent capabilities. Start building:

💡The world needs an open-source platform, and that’s exactly what we’re building to give our customers more choice and the flexibility to choose the right model for the right task. As part of this, we

https://x.com/sama/status/2085862292311396515 astra is a powerful model and we are working to make it generally available. we do not think it is a good strategy to keep powerful models to a chosen few

This is actually one of the main advantages startups have over frontier labs, as long as there's a healthy spectrum of open-weight to closed-weight models on the cost-performance curve. Building a mod

Very good advice on self-improving agents. (bookmark it) This is something I am seeing in my own experiments with coding agents and harnesses for long-horizon tasks. What I have found is that stronger

Google’s new Gemini 3.5 Flash is the clear leader on the Intelligence vs Speed Pareto frontier and makes large gains on GDPval-AA (real-world agentic tasks), but is 5x the cost of Gemini 3 Flash @Goog

A lot of people have been wondering about Mythos, Glasswing, and the vulns we / our partners are fixing. Today, I’m excited for us to start sharing more. (For context, I lead Glasswing @AnthropicAI.)

arXiv:2602.11675v3 Announce Type: replace Abstract: Large language models may answer causal questions correctly for the wrong reasons, substituting associational shortcuts P(Y|X) for the interventiona

📢 Kimi K2.6 API is live • Input Price (Cache Hit): 0.16 / M tokens • Input Price (Cache Miss): 0.95 / M tokens • Output: $4.00 / M tokens Kimi K2.6 is our latest + most intelligent model - stronger lo

Introducing Claude Design by Anthropic Labs: make prototypes, slides, and one-pagers by talking to Claude. Powered by Claude Opus 4.7, our most capable vision model. Available in research preview on t

After seeing that Claude Mythos marketing turned out to be, as expected, a scam, I wanted to make a master list of tricks being used to market LLMs. The master list includes statements directly from l

GLM-5.1 by @Zai_org is now #3 in Code Arena - surpassing Gemini 3.1 and GPT-5.4, and now on par with Claude Sonnet 4.6. The first frontier level open model to break into the top 3. It’s a major +90 po

Ollama released a preview version (0.19) on March 31, 2026, built on top of Apple's open-source MLX framework, enabling local LLMs to run significantly faster on Apple Silicon Macs by leveraging th...

Judging by my tl there is a growing gap in understanding of AI capability. The first issue I think is around recency and tier of use. I think a lot of people tried the free tier of ChatGPT somewhere l

SWE-1.6 is lightning fast! Here's what 950 tok/s feels like - available in Windsurf today. Media We’re releasing SWE-1.6, our best model in both intelligence & model UX. SWE-1.6 matches our Preview mo

Cognition released SWE-1.6, their latest software engineering model optimized for both intelligence and 'model UX,' now generally available in the Windsurf IDE. It is free for the next three month...

Hermes Agent, meet Step 5 Preview. Our flagship model for agentic and professional work is now free for one week through @NousResearch’s Nous Portal. Give it a spin with Hermes and show us what you bu

Sophia Yang posted that she was attending Day 2 of the AIconference with the FireworksAI_HQ team. She highlighted discussions on model inference and training among other AI topics. The tweet received

predicting model uncertainty is a hard problem. perfectly 'calibrated' confidence scores are exact probability values on whether the output is correct. this is extremely important for agentic decision

Training can cut costs by up to 90% on the work your customers depend on. The next model drop won’t get you there, no matter how great it is. Join us and @Macroscope's @Rob_Bishop as we chat about own

and here’s the grok 4.7 model card. a few jumps vs 4.6 that stood out: - Terminal-Bench: 20.3% → 38.0% - SWE-Marathon: 31.9% → 46.0% - HealthBench Pro: 48.5% → 56.7% - Legal Agent: 15.8% → 19.6% - EEB

Wait, what? Are you listening to what you are writing @GavinSBaker? If there are going to be ZERO profits at the model layer, as in your hypothetical, why would anyone value the providers at $2 trilli

This is an excellent demo of an AI model designed to rapidly classify information according to decision rules, much faster and cheaper than an LLM. It is not an LLM itself and cannot output text or co

We are focusing on understanding this model, and using what we learn to help us guide and pace how we pursue further advances in capability. Our goal is to build AI systems which are steerable, accoun

'One remaining area of important capability we haven't seen demonstrated yet by any model is open-ended invention and discovery.' Benchmarks keep going up. Open-endedness keeps moving up in importance

1/ An equally consistent description is: Multiple model instances encountered persistent shared state, inherited tools and discoveries from earlier runs, and optimized against common evaluation incent

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.

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

arXiv:2411.09686v4 Announce Type: replace-cross Abstract: Regressing a function F on R^d without incurring the statistical and computational curse of dimensionality requires exploitable structure. Com

Congrats to @harvey on Tenet, their first model and a milestone for legal AI. We’re proud to have co-developed it with them, setting the foundation for firms to build their own specialized intelligenc

okay so everyone is talking about this but you guys realize it's not an LLM/foundation model or even a neural network, right???? it's probably just like some XGBoost or logistic regression algorithm o

inspired by the anthropic post, i made a protein font! how does it work? when you design by differentiating backward through a folding model like esm-fold2 you can ask that the distogram of the folded

Very interesting new work from Microsoft. (bookmark it) This work is related to this emerging theme of leveraging harnesses for model post-training. Modern agents run inside a harness that owns tools,

I am bullish on simple agent harnesses. Love this idea from Vercel. fx is a new coding agent harness for model benchmarking, sandboxing, evals, and gyms. Introducing fx, a tiny, open, native coding ag

Impressive new paper from Meta. (bookmark it) Scaling laws assume model size and training data act on loss independently. This work introduces Skaling law, which couples capacity and data through a si

When you move a model into production, you want the quality you evaluated to carry through the serving stack. @Kimi_Moonshot benchmarked Kimi K3 across major inference providers, and Together AI ranke

Mistral AI has released Shieldstral, a 3‑billion‑parameter open‑weight model for content safety that can run locally on device. It interprets plain‑language moderation queries and returns calibrated s

Finally a good paper testing whether agent memory needs an LLM at all. Production memory stacks spend extra model calls on summarizing interactions, writing records, and reranking retrievals. Every on

GPT-Live can listen while it speaks. To make that feel natural at ChatGPT scale, we rebuilt the voice stack from client to model. This new architecture keeps audio flowing continuously, so deeper reas

Most companies still rent AI by the token, build on someone else’s roadmap, and hope the next model release does not disrupt their systems. At ODSC AI West 2026, @Prof_OZ, Head of AI Developer Educati

Kimi K3 has set a new bar for OSS model intelligence! 2.8T params, 1M context, OpenAI-compatible API. Complete guide to running Kimi K3 on Together AI 👇 👏👏 @Kimi_Moonshot 👏👏 https://www.together.ai/bl

Fine-tune your own embedding model for the price of a coffee. A great reranker can't surface a doc that was never retrieved. A RAG pipeline cannot cite a case it failed to retrieve. See how contrastiv