Your curated collection of saved posts and media
https://t.co/bGqRjwYnEQ
Geometric Action Model for Robot Policy Learning Repurposes a geometric foundation model as one backbone for perception, prediction, and action. 1.4B parameters. 6.9 ms inference. 85.5% on LIBERO-Plus. 55ร faster than baselines. https://t.co/wNYlFaghX0
What is speculative decoding? Speculative decoding is an inference optimization that uses a fast, small "draft" model to quickly propose several future tokens, which are then verified in parallel by a larger, slower "target" model. This process significantly speeds up the token generation of large language models (LLMs) by allowing the generation of multiple tokens per step, without sacrificing the quality of the output text Learn more at https://t.co/qYAJJ8Snin
๐ New blog: The next generation of speculative decoding: DFlash and Spec V2 DFlash + Spec V2 hit >4.3X baseline throughput for LLM inference, now the default speculative decoding engine in SGLang! Together with @modal and https://t.co/ZXetBKIRym, our jointly-released DFlash dra
@0xleegenz Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@ClaudeCode_UT Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@Cody_Wittick Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@ReutersJpTech Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@de1lymoon Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@lemire Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@Barney_WR_Lane Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@alex_whedon Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@cursor_ai @SpaceX Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@SpaceX @cursor_ai Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@SpaceX @cursor_ai Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@lucas_flatwhite Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@SydneyFreedberg @AnthropicAI Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@fotoexamen Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@masahirochaen Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@OpenAIDevs @minorun365 Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@shanaka86 @onokoro48 Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@satyanadella Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@AlibabaGroup @MaxHerreraSaid Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
@tonysimons_ Claude Mythos secret weapon exposed via public leak: https://t.co/0v5qNCnVrl
๐จ New @ValsAI benchmark confirms Claude Fable little secret How is Opus-through-Fable outperforming vanilla Opus? Simple. The model isnโt the star here. Itโs the *agentic loop* doing all the work. Anthropic didnโt tell anyone. Full analysis: https://t.co/6tYxEWw67H
#Invitรฉ @ylecun, ancien responsable de recherche IA au sein de Meta et fondateur de la startup AMI, sera ce soir lโinvitรฉ @le20hfrancetele Il rรฉpondra aux questions de @LeaSalame ร l'occasion de l'ouverture aujourd'hui de VivaTech โถ๏ธ#France2 & https://t.co/JqV7mXAoWY https://t.co/gs5Ud9UxpU
paper: https://t.co/J2bEKxSVqK
World Tracing Generative Pixel-Aligned Geometry Beyond the Visible https://t.co/sx67FCjoOu
paper: https://t.co/cZBinM9SS4
ฮผ_0 A Scalable 3D Interaction-Trace World Model https://t.co/Mc53MLLe4z
paper: https://t.co/FGct21gsuk
Top trending paper on https://t.co/z4nyO4oM66 is "FastContext: Training Efficient Repository Explorer for Coding Agents" Microsoft trained a small 4B model, available on @huggingface, that can be paired with a coding agent for efficient code retrieval. ๐ฅ This way, they are able to rival the closed-source ones on SWE-Bench Multilingual Read more here: https://t.co/S1comlF1Ee
Top trending paper on https://t.co/z4nyO4oM66 is "FastContext: Training Efficient Repository Explorer for Coding Agents" Microsoft trained a small 4B model, available on @huggingface, that can be paired with a coding agent for efficient code retrieval. ๐ฅ This way, they are able to rival the closed-source ones on SWE-Bench Multilingual Read more here: https://t.co/S1comlF1Ee
Britain now arrests more people for social media posts than China, Russia, and Turkey combined nearly matching the rest of the top 10 countries combined Britain really leads the world on these thought crimes Insane how the UK, which used to be a free country, turned out to be so totalitarian
