Name: LinkedIn · Ai Benchmark
Creator: LinkPost

Top posts sur ce sujet

C
Charafeddine Mouzouni
Applied AI for enterprises. Founder @ Cohorte, Professor @ OPIT.
21 juil. 2026
Chinese models changed the whole AI game. If you're not hosting them (or seriously considering it), you're probably burning money. Real numbers from real teams: an average coding team hits $20-30K a month on Claude. Th…
3520
LinkedIn
O
Obaloluwa Ola-Joseph Isaiah
Turn AI into your unfair advantage
20 juil. 2026
🚨 Moonshot AI has just unveiled Kimi K3, a new AI model that performs alongside the world's best models, including Fable 5 and GPT 5.6 Sol. Unlike those models, Kimi K3 comes with open weights. From July 27, develope…
34080
LinkedIn
C
Charafeddine Mouzouni
Applied AI for enterprises. Founder @ Cohorte, Professor @ OPIT.
21 juil. 2026
Chinese models changed the whole AI game. If you're not hosting them (or seriously considering it), you're probably burning money. Real numbers from real teams: an average coding team hits $20-30K a month on Claude. Th…
3270
LinkedIn
M
Maxime Labonne
Head of Post-Training @ Liquid AI
8 juin 2026
🏎️ Nemotron 3 Ultra: what distillation can't fix Nvidia released their biggest model to date, Nemotron 3 Ultra, a 550B MoE. I wrote an article on Substack analyzing this release, especially the post-training section. …
31514
LinkedIn
O
Obaloluwa Ola-Joseph Isaiah
Turn AI into your unfair advantage
11 juin 2026
Claude Fable 5 just dropped and within 24 hours someone asked it the most basic AI benchmark question. "𝘏𝘰𝘸 𝘮𝘢𝘯𝘺 𝘙𝘴 𝘢𝘳𝘦 𝘪𝘯 𝘵𝘩𝘦 𝘸𝘰𝘳𝘥 𝘚𝘵𝘳𝘢𝘸𝘣𝘦𝘳𝘳𝘺?" We built AGI and it has a superiority comp…
265128
LinkedIn
M
Maxime Labonne
Head of Post-Training @ Liquid AI
21 mai 2026
⚓ Anchored Supervised Fine-Tuning ASFT is a new post-training method that extends Dynamic Fine-Tuning (DFT) with a KL regularization term to the base model. The authors claim it closes most of the gap with full RL at SF…
2426
LinkedIn
E
Ethan Mollick
Associate Professor at The Wharton School. Author of Co-Existence, coming October 20!
29 juin 2026
I took the new AA-Briefcase scores from Artificial Analysis (basically having the AI do multi-week consulting gigs with a lot of complexity) and graphed the frontier curve for open and closed models: 1) Surprise, rapid g…
20931
LinkedIn
M
Md Riyazuddin↗️
LinkedIn Top Voice • AI Enthusiast • Personal Branding • Helping brands to grow 📈 • Data Science • DM 📩 for collaboration
11 juin 2026
Anthropic just opened the doors to something that was never meant for the public. Claude Fable 5 is now available. And the results are wild. Hyperagent tested it on 5 open-ended projects that most AI models struggle t…
20616
LinkedIn
E
Ethan Mollick
Associate Professor at The Wharton School. Author of Co-Existence, coming October 20!
16 juil. 2026
My benchmark where I have AIs create one file procedurally-generated harbor towns through history in one shot now has GPT-5.6 Pro, Fable, Kimi K3, and Inkling. You can play with all the simulations: https://lnkd.in/e8x4…
16224
LinkedIn
E
Ethan Mollick
Associate Professor at The Wharton School. Author of Co-Existence, coming October 20!
17 juin 2026
I have a fun, oddly useful AI benchmark: "build me a procedurally generated 3D simulation showing the evolution of a harbor town from 3000 BC to 3000 AD, it should look beautiful & allow me to have some control over it" …
13929
LinkedIn

LinkPost - Tendance LinkedIn : Ai Benchmark · analyse, évolution & top posts | LinkPost

Une question ?

Salut !

Une question ?

Salut !

Ai Benchmark

Évolution sur 12 semaines

Top posts sur ce sujet