์ ๋ฒ์ฃผ์ ์ ค๋ค ํ๋ค๊ฐ ์ด๋ฒ์ฃผ๋ GTA ํ๋๊น ๋์ ์ฐ๋ น์ ์ฐจ์ด๊ฐ ๋๋ฌด ์ ๋ช
ํจ
18
WKBae ๐ฏ๏ธ retweeted
์ ์๋ฌด๋ Neon์ผ๋ก Neon ๋ง๋ค๊ธฐ ๊ฐ์ ๊ฑธ ์ํ๋๊ฑฐ์ง
Gemini 4 Neon ยท xhigh | 48min | 1st attempt
WKBae ๐ฏ๏ธ retweeted
ํธ๋ผํ, ๋ฏธ๊ตญ ๋ด ์์คํ
์ฅ์ ์์ถ์ "๊ด์ฐฎ๋คโฆ๊ทธ๊ฒ ์ธ์"
yna.co.kr/view/AKR2026111802โฆ
The harness is a crucial lever on model behavior. Compared with a Bash-only toolkit, native harnesses generally raise both solve rates and costs.
However, this effect isn't uniform. In Codex, 6.1 Terra and 6 Luna gain 10 points, while the newer 6.2 models see little change in solve rate. Grok and DeepSeek make big gains from their harnesses, while Muse stays flat. Unlike others, Fable 5.2 is much cheaper in its native harness.
์๋ 50%๋ฅผ ํ ์ธํด์ฃผ๋ Flex ๋ชจ๋๊ฐ GPT-6.2๋ถํฐ ์ฌ์ฉ ๋น ์ก๊ตฌ๋
Not every workflow requires an immediate response.
For single- or multi-turn workflows that can wait, save up to 40% per token than Standard on GPT-6.2 Astra or GPT-6.2 Sol.
In Responses API background mode, choose a 2h or 8h target for the first response. These discounted windows are now in public beta.
1
1
20
๋ชจ๋ธ ๋ช ๊ฐ๋ฅผ ๋ด๋๊ฑฐ์ง
Introducing Gemini 4 Neon! Pushing the frontier across domains and punching above its model class in a meaningful way. To me, Neon feels like the right balance of intelligence, speed, and usability. Available now across @GoogleAIStudio, the API, and more!
40
WKBae ๐ฏ๏ธ retweeted
์๊ธฐ๊ฐ ์ปจ๋ํ ๊ธฐ๊ณ๋ ์๊ธฐ๊ฐ ๋ง๋ค์ด์ผ์ง ์ฌ๋ค์ฐ๋ ๊ฑฐ ๋ณด๋ฉด ์ด ๋๋ผ ์ ์๊ณตํ์ ๋ฏธ๋๊ฐ ์ฐธ ์ด๋ก์ง์ด
Flex๋ ๋น๋ฐ๋ฒํธ ๋ก๊ทธ์ธ์ ์์ ์์ ๋๊ตฌ๋
guide.flex.team/ko/articles/โฆ
22
Anthropic์ ํ๋ฅด๋ง ์ ๋ฆฌ ํ์ํํ๋ค๊ณ ์์ด์ ํธ 1,000๊ฐ๋ฅผ ๋ณ๋ ฌ๋ก 28์๊ฐ ๋ด๋ฆฌ ์คํํ๊ณ OpenAI๋ ๋๋น์-์คํก์ค ๋ฌธ์ ํ๋ ค๊ณ ์์ด์ ํธ 10,000๊ฐ๋ฅผ 88์๊ฐ๋์ ๋๋ ธ๋ค.
์๋ฌด๋ฆฌ ํ ํฐ๊ฐ์ด ์ ๋ ดํด์ง๋ค ๋งํด๋ดค์ ์์ ํ์ฌ์ ๋
์ ํ์ ๋์ ๋ด๊ณ ์ฌ์ฉํ๋ ๊ตฌ์กฐ์์ ๋ถํ๋ฑ์ ์ฌํด์ง ๊ฒ์ด๋ค.
AI๋ ๋๊ตฌ์๊ฒ๋ ์ด๋ ค ์๋ ํ๋ฑํ ๊ธฐ์ ์ด ์๋๋๋ค. AI ์จ๋ณด์
จ์ผ๋ฉด ์ ํ
๋ฐ, ๋ ๋ง์ ํ ํฐ์ ํ์ธ ์ ์๋ ์ฌ๋์ด ์ด๊ธฐ๋ ๊ตฌ์กฐ์์. ๊ณ ์ ๋ฌด๋ฃ ๋ฒ์ ์ ์ธ ์ ์๋ ์ฌ๋๊ณผ, ์์ญ๋ง ๋ฌ๋ฌ์ ํ ํฐ์ ํ์ธ ์ฌ๋ ฅ์ด ์๋ ์ฌ๋์ด ๋ด๋ ๊ฒฐ๊ณผ๋ฌผ์ด ๋ค๋ฆ
๋๋ค.
Meet Gemini 4 Hydrogen.
It brings cutting-edge intelligence to everyone, with speed and efficiency for everyday coding and agentic work.
Across 4 thinking levels, it stands on its own or teams up with Carbon and Argon โ 2.5X faster than 3.8 Flash, outpacing competing models in its class.
WKBae ๐ฏ๏ธ retweeted
introducing the Gemini Decisions API
powered by Gemini Decision 1, our new multimodal decision model
get fast, calibrated decisions on up to 1000 questions at once across text, images, audio, video, and PDFs
try it today in AI Studio and via the Gemini API ai.studio
๋๋์ด ๋ญ ์ก๋ ์นผ์ด ๋์ค๋๊ฑด๊ฐ
Introducing e-means, an algorithm beating k-means in 10x fewer epochs.
With order of magnitude wall-clock speedups, e-means is released in FAISS and needs no hyperparameter tuning beyond k. It is robust to various scales and datasets. Still skeptical ? Just try e-means [1/10]๐งต
12






























