#openmodels

Guillermo Flor @guilleflorvs

BREAKING: Revolut moved ~2,000 engineers onto self-hosted GLM for coding and pushed Claude Code and Codex to the exception list. Per David Abitbol, who leads AI ops there: 56 Nvidia B200s, a blended 30 cents per million tokens, roughly a tenth of what they paid Anthropic and OpenAI. Engineering lead time improved 70%. Here's the part that matters: the gains showed up when tokens stopped being rationed. Engineers quit saving AI for the hard tickets and pointed it at everything. Frontier models become the exception, open weights become the default. That's a pricing problem for every lab heading to IPO. h/t @OliverMolander Would you move your team off frontier APIs if the bill dropped 90%?

Cline @cline

We've kept hearing how GLM-5.2 beats Opus 4.8, and are skeptical of benchmarks - so we tested them on a real bug from the Cline repo. While both models fixed the issue, GLM was the winner in terms of cost and code quality: - GLM used twice as many tokens (GLM 1.1m vs Opus 660K) but cost half as much (GLM $0.41 vs Opus $0.81) - Opus finished quicker - 1.6 min and 12 tool calls vs GLM 4.7 min and 28 tool calls - GLM cleaned up dead code and verified the build compiled before completing. Opus didn't - it left type errors that passed tests but broke the production build. Both runs used the same Cline harness prompting and tools, so it seems GLM is RL trained to spend more tokens verifying its work before completing. Impressive work by the @Zai_org team!

Kelly Wright @mrkellywright

@burrytracker Their big problem is that big companies with $ have figured out what Palantir has. Open models, that you own and control are the future.

Alex Imas @alexolegimas

The longer I’ve spent time with this paper the bigger of a deal it seems. The economic implications are quite significant. This is a frontier expert task. This is Qwen3-235B.

Arnaud Bertrand @RnaudBertrand

Eric Schmidt saying the quiet part out loud: "What I don't like about [China's AI] is that it's all open source which means it's largely uncontrolled and not controlled in any way by us." He adds, "if that makes you feel any better," that only 2 or 3 countries can be independent AI powers. In other words, it's all about hegemony: the ideal scenario is a world where AI is controlled by the US - and the fewer countries that can resist that, the better. Src for the video:

Dharmamitra @dharmamitra_ucb

The Dharmamitra team is very proud to announce that we have switched our translation engine to a self-trained model recently, which in LLM-judge based evaluation outperforms our previous Gemini-based pipeline as well as Claude Opus consistently. 1/

Sentient @SentientAGI

SENTIENT RESEARCHER SPOTLIGHT: Pramod Viswanath A cofounder of Sentient and Professor of Electrical & Computer Engineering at @Princeton, @viswanathpramod’s research spans information theory, cryptography, and learning theory, with a focus on making open models verifiable, attributable, and controllable. He helped define the OML paradigm—ownership, control, and alignment for open models—and co-led OML 1.0 (Fingerprinting), demonstrating large-scale, utility-preserving model attribution. Previously at UIUC and now at Princeton, he’s widely recognized across academia for foundational contributions in information theory and cryptography—and for translating them into practice for open-source AI. Beyond OML, he has driven rigorous, field-scale evaluations and formalized governance primitives that tie cryptographic accountability, licensing economics, and alignment into a single fiduciary-first framework for open models.

David Romero @Davidromogr

Excited to introduce Q-ViD, a simple, 𝐨𝐩𝐞𝐧, and effective approach for video QA, that achieves competitive or even higher performance than prior works based on more complex architectures or 𝐜𝐥𝐨𝐬𝐞𝐝 GPT models. 🧵 w/ @thamar_solorio