Click any tag below to further narrow down your results
Links
This arXiv paper claims an open 27B model can diagnose its own failures and edit its own weights, with no gradients and no training, and end up measurably more accurate. The authors frame it as "self-rewriting" and a possible step toward recursive self-improvement.
- The model's own edit raised its accuracy 6.6 points on problems it never saw during the edit.
- Across eight consecutive sessions it tested edits on practice problems and submitted only ones that worked, and accuracy only went up.
- Given a damaged copy of itself, it restored that copy to over 80% of original accuracy on unseen image tasks, without reading the original weights.
This is the sign-in page for Liquid AI's chat playground, where you can talk to the company's own family of models. The lineup spans text, vision, and voice, with the pitch being strong performance at much lower compute cost than typical frontier models.
- The lineup includes LFM2.5-24B-A2B and LFM3-1.7B for text, LFM3-VL-2B for vision, and LFM3 Omni 2B and LFM3 Audio 2B for voice.
- There is also LFM2.5-1.2B-JP, a text model aimed at Japanese.
- The page offers no benchmarks or pricing, so the efficiency claim is unverified from this text alone.
@thesupermannx on X
A Meta paper called "Breaking the Token Ceiling" argues that language models can skip the tokenizer entirely and train on raw bytes, a vocabulary of just 256 values. The author claims the distilled byte models match or beat token-based models like Llama 3.2-1B and Gemma-3-1B while needing less data and storage.
- Current models like ChatGPT, Claude, and Gemini split text into tokens drawn from a vocabulary of roughly 100,000 chunks.
- Meta's paper trains models on raw bytes with a 256-value vocabulary instead of tokens.
- Distilled 1B-parameter byte models reportedly asymptotically surpass Llama 3.2-1B and Gemma-3-1B on downstream tasks.
Anthropic released Claude Haiku 5.5, a small language model priced 75% lower than its predecessor while delivering significantly better performance on tasks like summarization, customer support, and coding. The company also cut Sonnet 5.5 cache read prices in half and added monthly API credits for paid subscribers.
- Haiku 5.5 costs $0.10 per million input tokens (vs. $1.00 for Haiku 4.5) and outperforms Haiku 4.5 across benchmarks—scoring 72.4% on OSWorld computer use tasks compared to 15.7% for the older model.
- The model includes an adjustable effort setting that lets users trade off speed for accuracy, and is positioned as ideal for high-volume, cost-sensitive work like batch summarization, database queries, and subagent tasks in larger AI systems.
- Sonnet 5.5 cache reads now cost 50% less ($0.10 instead of $0.20 per million tokens), reducing agentic task costs by ~20%, while Max and Team subscribers get monthly API credits ($100–$500) to experiment with building agents and applications.