La course à la Super IA est violente @ericschmidt dit que l’IA sera plus intelligente que la somme des cerveaux humains en 2031 Zuckerberg vient de débaucher le patron des IA de @Apple avec 200 millions de dollars par an Qui préférez-vous comme vainqueur de cette guerre ?
GENERATIVE AI
-
Flux LoRA for Creating Aged Damaged Images via Replicate
By
–
A flux kontext dev lora to make an image old and damaged, using Replicate's prototype trainer.
-
Grok’s Value Conflicts and AI Alignment: The HAL 9000 Parallel
By
–
The whole Grok situation (system prompt changes with values that conflict with post-training and pre-training values) is, oddly enough, similar to the reason the fictional AI HAL 9000 went insane, as was revealed in 2010, the sequel to 2001
-
LFM2 Model Now Available on LM Studio Beta
By
–
LFM2 is now available on @lmstudio 🎉
— Maxime Labonne @ ICLR (@maximelabonne) 13 juillet 2025
You can play with it today using the beta channel.
I noticed that other small models often struggle to provide relevant answers within a structured output. We trained LFM2 on so much structured outputs that it doesn't mind at all. pic.twitter.com/fn1egMXsZBLFM2 is now available on @lmstudio You can play with it today using the beta channel. I noticed that other small models often struggle to provide relevant answers within a structured output. We trained LFM2 on so much structured outputs that it doesn't mind at all.
-
Midjourney celebrates three years of AI image generation
By
–
It's been 3 years since we first made Midjourney available to everyone. Thank you for being a part of our story so far. We're privileged to explore this vast, unimaginable, beautiful future together <3
-
Kimi Now Available Through Anthropic API Integration
By
–
Very interesting – you can use Kimi with the Anthropic API. This means, perhaps most importantly, that you can now use Kimi with Claude Code!
-
Recorded demo showing Grok 4 Heavy behavior
By
–
You've been aggressively rude and I just recorded a 5 minute video for you of it working in Grok 4 Heavy. The discrepancy between normal Grok 4 and G4H is explicitly discussed in the thread.
-
Replications and Grok 4 Heavy chat shares
By
–
There are several replications from others with Grok 4 Heavy access in the replies. I also included Grok share links (with no custom instructions) in the thread for five separate chats repeating this prompt in Grok 4 Heavy.
-
Scaling Laws and Tool Use Driving AI Progress Forward
By
–
Grok 4 suggests that scaling still works (with the diminishing returns predicted by the scaling law), and that tool use can unlock performance gains. Kimi suggests there continues to be big opportunities from improvements in methods (Muon, etc.). Lots of paths for AI right now.