Search: “models”

LightOn OCR-3 is now available on OpenDocRouter, priced at 0.28 / 1m input and 1.40 / 1m output (it is ~$3.19 / 1k pages on ParseBench). We benchmarked it on ParseBench. It is on the pareto frontier f

Meta published a paper that might end the transformer era. For the last seven years, every major AI, ChatGPT, Claude, Gemini, has been built on the exact same architecture. Transformer. But Transforme

A 200 AI subscription can be worth 12,000 in tokens. Which plan delivers it is not close. 'They're selling to willing buyers at the fair market price. This is capitalism.' 'You can pay 200 for a Claud

OpenAI’s tweet (posted Oct 7 2026 at 6:05 PM) promotes the idea that visual content can deepen comprehension of a topic, highlighting how images make concepts “come to life.” The post, part of OpenAI’

OpenDocRouter is here! https://www.opendocrouter.ai/ Some time ago, I realized LlamaIndex was really good at benchmarking and applying nearly every model to document parsing (see: ParseBench). So, why

The Windows launch event today has mentioned 'local AI' 41 times (so far) 🤯 Media It is very cool to see llama.cpp on the big stage in todays Windows event! The software and hardware stacks are finall

Grok 4.7 is excellent at dealing with large code repositories Grok 4.7 takes the TOP 3 spots on VulcanBench Frontier v4, outperforming Fable 5.1, Opus 5.5, GPT-6 Astra, GPT-6.1 Sol and other leading f

Mistral can pelican now! https://tools.simonwillison.net/markdown-svg-renderer?url=https%3A%2F%2Fgist.github.com%2Fsimonw%2F0db8b8196e57711b352f6c5857cf5b35#response-1 Meet Mistral Large 4, aka Le Cho

Mistral Large 4 against the field: efficiently reasoning over diverse complex challenges. CTF speedrun: 19 challenges, with tool calls and solve times grounded in actual runs. It solved 18 of the 19.

I actually want personal operating systems optimized for co-working with AIs, including fine-grained ideas of security, easier ways to dynamically see & understand what AIs are doing, and independent

Super interesting paper from Meta Superintelligence Labs on controlling long agent runs. They use the same workers and same budget, and ProgramBench goes from 63.7% to 71.5% with GPT-5.5 when a dedica

Banger paper from Meta and colleagues. (bookmark it) They discuss benefits of giving your agent write access to its own context. In other words, they investigate how effective it is to allow a languag

In light of the latest Anthropic's latest not-particularly-oblique attempt to build a case against open weights, I thought I would write down how I currently think about open weight AI models. (1) Ope

Independent numbers from @ValsAI. Fewer tokens per turn means lower cost and faster agent loops on every call. Try Ember-1 on Fireworks Serverless: https://fireworks.ai/models/fireworks/ember-1 Firewo

Perplexity AI released a preview of its embedding model “pplx‑embed‑v2‑context‑9b” on Hugging Face on September 30, 2026 (7:07 PM local time). The Hugging Face repository page displays 3 122 views and

Hugging Face Jobs usage is stonking 📈 All thanks to agents: you just need the hf CLI installed and you have access to an insane amount of hardware. Asked mine to test hundreds of code snippets from Hu

Usage in Devin counts toward the Codex and ChatGPT Work usage already included in your plan. You can let Devin use all of it, or set a lower limit for Devin in your settings. OpenAI models are part of

who's got two thumbs and just ranked #6 of 49 on image-jevbench? 👍 this guy 👍 one of the fastest in top 10, accuracy is close. calibration of confidence is where it's weakest. @airesearch12 says a sim

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

Everyone thinks Claude is just good at writing code and hacking critical infrastructure. Few people know that Claude is also an accomplished artist, filmmaker, editor, cinematographer and art director

Reliable, fast inference is essential. As Ted shared at Apsara Conference 2026, Together makes it easy for teams to get access to frontier open models without needing to manage any of the infrastructu

Remember October 2022 when you doused Galactica with vitriol? Galactica was a 120b-parameter LLM-based system from Meta-FAIR designed to help scientist write papers. It was open sourced (link below).

Proud to be a launch training partner for @LangChain's LangSmith Fine-Tuning and smithtune. LangSmith traces → managed SFT on Fireworks → eval → deploy. One CLI, no training infra. Get a Fireworks API

At Box, we've been testing Opus 5.5 on a variety of complex enterprise knowledge work tasks dealing with unstructured data with the Box Agent. Overall, we saw frontier capability levels, with major pe

Claude Opus 5.5 is now available as the Standard effort level in Perplexity Computer. On our WANDR benchmark, it scored 0.610 at $4.13 per task, slightly outperforming Claude Fable 5.1 while costing 6

In our evaluations, we see that GPT-6 Sol reads about 17% less context than GPT-5.6 Sol to reach the same score, and rarely leaves a repo with failing tests after its last edit. GPT-6 Luna’s biggest g

AI can be a really wonderful tool for exploring topics far from coding. I had Claude Fable 5.1 put together an annotated guide to Eliot's poem 'The Wasteland,' with multiple pathways through the poem,

Found a great production use case for Jev. I used Jev to organize ~2.3K AI research papers. The total cost was $0.14, and it took about 83 seconds. The process: The papers already had old tags, which

Grok 4.7 places @SpaceXAI as third, after Anthropic & OpenAI, for agentic coding. When factoring in that Grok is significantly faster & lower cost, it’s a great choice for your everyday workhorse. Gro

Thanks @sgl_project for the day-0 support! 🙌 SGLang-Diffusion now serves Qwen-Image-2.1: text-to-image generation, multi-image editing, and transparent RGBA output. Try it out! 🎨 Day-0 support for @Al

Epoch continues to do some of the best public benchmarking work on AI. This is helpful (and tells us how bad the state of benchmarking is, and how terrible some of our favorite benchmarks are) Introdu

Why @CompleteSkeptic created ChatGPT and then spent 2 years in stealth building Jev Media After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent t

Come hang out with AI engineers, tech leads, and the Fireworks team during AI Engineer 2026. Want to leave with hands-on experience fine-tuning, evaluating, and deploying a model on Fireworks? Bring y

Important paper on improving agent coordination. On normal tasks, this work shows agents caused harm 0 to 5% of the time. After receiving an unsafe trajectory from another agent, that rose to 40 to 95

Nice work from Adobe. It's standard practice to store a fixed reference answer for every eval case. When the underlying data changes daily, that stored answer goes stale. Adobe researchers write each

This film cost 29,575 and took 10 days to make. AI ≠ Cheap Our Nexus feature will still cost millions next year. But it'll look like a 200M film. I have a Claude skill you can copy that makes this eas

Gemini 3.8 Live is rolling out to: — Consumers: in Search Live — Developers: in public preview in the Gemini API via @googleaistudio — Enterprises: in private preview via Gemini Enterprise (coming soo

Well.. @levie and I filmed this episode of Training Data a week or two ago, when the “current thing” was Doug Leone’s novacaine root canals instead of pacing the frontier… Simpler times! But Aaron’s a

On September 14 2026 (02:41 UTC) Teknium announced that **ChatGPT‑Image‑2.5** is now available on the **Nous Portal**, enabling it to be integrated into the **Hermes Agent**. The model can also be acc

Boris Cherny announced on September 14, 2026 that the software *Fable* had cracked the 370‑year‑old Cyphral Distich cipher. The claim was posted via Vals AI’s platform and highlighted a demonstration

I have conducted an audit of Anthropic's finances. What I have found is so shocking that I am calling for a Congressional investigation. Anthropic is not just seeking regulatory capture. It has built

Law enforcers already have authority to charge companies and their CEOs for creating and releasing dangerous, unvetted, or defective products. We shouldn’t let discussions about new legal regimes dist

In his Pacing AI letter, Dario claims that RSI is starting to happen. The problem with this statement is that both the blog posts he links to *and* the system cards for the latest Claude and ChatGPT m

AI Alarm Hype Cycles - A brief history GPT-2: OpenAI withholds weights due to misuse/misinformation concerns GPT-3: No open release, Dario Amodei and other GPT-3 team members depart OpenAI to found An

together fine-tuning now supports glm-5.3, kimi k2.7-code + more we also added live run metrics, experiment comparisons, early stopping, dataset previews + finer training controls prices are down 30–7

OpenAI announced that its Premium plan now includes financial datasets from Daloopa, PitchBook, and LSEG News. This upgrade expands the premium offerings with up‑to‑date market, company, and news data