Mastodawn

Fili (@filiksyos)

더 무겁고 비싼 '사고하는' 모델들에 지쳤다는 내용의 트윗으로, 빠르고 저렴한 '비사고(non-thinking)' 모델을 원한다고 밝힙니다. 예시로 Composer 1과 Kimi K2.5를 언급하며 자신의 'openclaw'에 쓸 수 있는 가벼운 모델을 요구하고 있습니다.

https://x.com/filiksyos/status/2029658989898637491

#modelefficiency #edgeai #composer #kimi

Fili (@filiksyos) on X

So tired of more expensive, heavy, slower, thinking models Give me fast, cheap, non-thinking model that i can use for my openclaw Like composer 1, kimi k2.5

X (formerly Twitter)

AI Daily Post Feb 26

New research shows how speculative decoding trains a draft model to guess tokens, then verifies them with the main LLM—cutting compute and boosting token generation speed. The approach promises big gains in model efficiency and opens doors for open‑source AI training. Dive into the details! #SpeculativeDecoding #TokenGeneration #ModelEfficiency #OpenSourceAI

🔗 https://aidailypost.com/news/speculative-decoding-trains-drafter-guess-verify-llm-outputs

AI Daily Post Feb 26

Alibaba just released the Qwen‑3.5‑Medium model as open‑source, delivering Sonnet 4.5‑level performance on a single GPU. It uses a Mixture‑of‑Experts architecture and a new “Thinking Mode” to boost AI inference efficiency while staying lightweight. Dive into the details and see how this could reshape open‑source LLM development. #Qwen3_5 #OpenSourceLLM #MixtureOfExperts #ModelEfficiency

🔗 https://aidailypost.com/news/alibaba-open-sources-qwen35-medium-models-sonnet-45-performance

sayzard Feb 20

ARC Prize (@arcprize)

Gemini 3.1 Pro가 Google DeepMind의 ARC-AGI 세미프라이빗 평가에서 성능을 공개: ARC-AGI-1에서 98% 달성(과제당 비용 $0.52), ARC-AGI-2에서 77% 달성(과제당 비용 $0.96). Gemini 계열이 성능과 비용 효율성의 파레토 프론티어를 밀어붙이고 있음을 시사.

https://x.com/arcprize/status/2024522812728496470

#gemini #arcagi #benchmark #modelefficiency

ARC Prize (@arcprize) on X

Gemini 3.1 Pro on ARC-AGI Semi-Private Eval @GoogleDeepMind - ARC-AGI-1: 98%, $0.52/task - ARC-AGI-2: 77%, $0.96/task Gemini to push the Pareto Frontier of performance and efficiency

X (formerly Twitter)