*actually route everything to Claude*
@maximelabonne
-
Kimi K2.6 Routes Tasks Across Multiple Specialized Agents
By
–
Claw Groups adds another layer to the agentic madness: Kimi K2.6 routes different tasks to different agents with different capabilities Built on top of Agent Swarm. I don't know if it's really useful, but it sounds so cool
-
Tokenizer Switching: Computational Efficiency in LLM Operations
By
–
It's possible but switching to a new tokenizer is actually very cheap and easy, so it'd be a big waste of compute
-
Mythos High Score Graphwalks BFS: Training Methods Analysis
By
–
iirc this theory is based on Mythos's high score on Graphwalks BFS. There are many easier ways to explain this score (data, RL, distillation), so I don't think it's a strong argument to explain it
-
Mythos Tokenizer Switching Opus Through Midtraining Distillation
By
–
My bet is that Mythos uses a new tokenizer, and they switched Opus over to it (through midtraining) for distillation
-
Deep Dive Learning Guide to Data Structures and Algorithms
By
–
Excellent deep dive to learn about DSA
-
Never Too Late for Cheeky Kimi K2.5 Fine-tuning
By
–
It's never too late for a cheeky Kimi K2.5 fine-tune
-
Liquid AI Releases 34 Slides on Edge Model Design Training
By
–
I'm releasing the 34 slides on how we design and train best-in-class edge models at @liquidai I presented these slides yesterday at @aiDotEngineer They cover model architecture, pre-training, scaling laws, post-training, and even a solution to fix doom loops Special thanks to
-
LFM2.5-VL-450M Vision-Language Model Now Available
By
–
Available today on @huggingface! huggingface.co/LiquidAI/LFM2…