正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Introducing FlashCompact - the first specialized model for context compaction 33k tokens/sec 200k → 50k in ~1.5s Fast, high quality compaction

Morph

3,332 subscribers

226,726 次观看 • 5 个月前 •via X (Twitter)

教育新闻政治科学技术

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

LLMs have a limited context window. When conversations grow too long this affects output quality, performance and cost. New blog post from Earendil engineer Vegard Stikbakke on how compaction addresses this and how we’ve implemented it in Pi. Read the full post below

LLMs have a limited context window. When conversations grow too long this affects output quality, performance and cost. New blog post from Earendil engineer Vegard Stikbakke on how compaction addresses this and how we’ve implemented it in Pi. Read the full post below

Pi

258,965 次观看 • 19 天前

Introducing Seedream 3.0, available now in the Freepik AI Suite Crazy prompt adherence, blends styles with precision, and gives fast, high-quality results Images made with this new model 🧵👇

Introducing Seedream 3.0, available now in the Freepik AI Suite Crazy prompt adherence, blends styles with precision, and gives fast, high-quality results Images made with this new model 🧵👇

Magnific

34,205 次观看 • 1 年前

Grok Code Fast 1 just shattered records on OpenRouter, crossing 237B tokens in a single day. The first model ever to move past the 200B/day milestone.

Grok Code Fast 1 just shattered records on OpenRouter, crossing 237B tokens in a single day. The first model ever to move past the 200B/day milestone.

DogeDesigner

60,691 次观看 • 9 个月前

introducing Seedance 2.0. the world's most powerful video model, supporting text, image, video, and even audio as input, and capable of producing long multi-shot videos with high quality. now available for everyone.

introducing Seedance 2.0. the world's most powerful video model, supporting text, image, video, and even audio as input, and capable of producing long multi-shot videos with high quality. now available for everyone.

Krea

44,876 次观看 • 4 个月前

First steps for a specialized DeepSeek v4 Flash inference engine focused on inference quality / stability at different quantizations, with networked API that is batching capable. This is the 2 bit quants model running on my M3 Max 128GB.

First steps for a specialized DeepSeek v4 Flash inference engine focused on inference quality / stability at different quantizations, with networked API that is batching capable. This is the 2 bit quants model running on my M3 Max 128GB.

antirez

14,220 次观看 • 4 个月前

Introducing our most accurate /search yet. We trained a model to return the excerpts that best answer your query, giving agents highly relevant context from each result. It's SOTA on SimpleQA while using 10x fewer tokens than processing full pages. Now on by default for free.

Introducing our most accurate /search yet. We trained a model to return the excerpts that best answer your query, giving agents highly relevant context from each result. It's SOTA on SimpleQA while using 10x fewer tokens than processing full pages. Now on by default for free.

Firecrawl

35,960 次观看 • 1 个月前

GROK 4 FAST 2M CONTEXT – NOW IN YOUR POCKET A reasoning model with a 2M context window — rewriting the cost curve of intelligence! Here’s the Hit List: •⁠ ⁠Free on iOS and Android •⁠ ⁠Multimodal brain built for more than text •⁠ ⁠2M context leaves competitors wheezing Source: xAI

GROK 4 FAST 2M CONTEXT – NOW IN YOUR POCKET A reasoning model with a 2M context window — rewriting the cost curve of intelligence! Here’s the Hit List: •⁠ ⁠Free on iOS and Android •⁠ ⁠Multimodal brain built for more than text •⁠ ⁠2M context leaves competitors wheezing Source: xAI

Mario Nawfal

84,937 次观看 • 11 个月前

(Z.ai) is now a model provider inside TRAE. You can BYOK and add GLM-4.6 and GLM-4.5 to TRAE IDE. GLM-4.5 supports up to 128K context and GLM-4.6 expands up to 200K context. Both deliver strong coding and reasoning performance on the benchmark and offers dual “thinking/fast” modes. Available now in the latest version.

(Z.ai) is now a model provider inside TRAE. You can BYOK and add GLM-4.6 and GLM-4.5 to TRAE IDE. GLM-4.5 supports up to 128K context and GLM-4.6 expands up to 200K context. Both deliver strong coding and reasoning performance on the benchmark and offers dual “thinking/fast” modes. Available now in the latest version.

TRAE

63,585 次观看 • 10 个月前

💰Introducing the Initial Labor Offering (#ILO) for Autonomys | AI3.0 🗓 Timeline: March 24th to April 24th, 2025 - 240 Social Miners and invited KOLs pledge work, submit deliverables, and receive rewards in $AI3 tokens based on quality, originality, and engagement. 🥇High-quality contributions and strong social media engagement will yield sizable $AI3 allocations 🎆Showcase your skills and let's create something extraordinary together!

💰Introducing the Initial Labor Offering (#ILO) for Autonomys | AI3.0 🗓 Timeline: March 24th to April 24th, 2025 - 240 Social Miners and invited KOLs pledge work, submit deliverables, and receive rewards in $AI3 tokens based on quality, originality, and engagement. 🥇High-quality contributions and strong social media engagement will yield sizable $AI3 allocations 🎆Showcase your skills and let's create something extraordinary together!

DAO Labs

10,198 次观看 • 1 年前

Introducing your.MOV - The First AI Native Vibe Video Editor 🔥 Edit Videos like A PRO using AI Prompts 👇 🧙‍♂️ Understands the footage context 🤖 Make edits using a chat-like interface ⚡ Add captions automatically 💖 Download high-quality output Try it now: #sponsored #ad Explore how to get started below 🧵👇

Introducing your.MOV - The First AI Native Vibe Video Editor 🔥 Edit Videos like A PRO using AI Prompts 👇 🧙‍♂️ Understands the footage context 🤖 Make edits using a chat-like interface ⚡ Add captions automatically 💖 Download high-quality output Try it now: #sponsored #ad Explore how to get started below 🧵👇

Madza 👨‍💻⚡

37,570 次观看 • 1 年前

Introducing 📦𝗔𝗿𝘁𝗶𝗟𝗮𝘁𝗲𝗻𝘁🔧 (SIGGRAPH Asia 2025) — a high-quality 3D diffusion model that explicitly models object articulation, paving the way for richer, more realistic assets in embodied AI and simulation: – Generates fully articulated 3D objects – Physically plausible joints & motion – High-fidelity 3D Gaussian appearance – Supports generation from a single real image arXiv: Project: Code (coming soon):

Introducing 📦𝗔𝗿𝘁𝗶𝗟𝗮𝘁𝗲𝗻𝘁🔧 (SIGGRAPH Asia 2025) — a high-quality 3D diffusion model that explicitly models object articulation, paving the way for richer, more realistic assets in embodied AI and simulation: – Generates fully articulated 3D objects – Physically plausible joints & motion – High-fidelity 3D Gaussian appearance – Supports generation from a single real image arXiv: Project: Code (coming soon):

Xingang Pan

11,520 次观看 • 9 个月前

LTX-2 is THE open source video AI moment I've been waiting for--the first to generates high quality video AND is FAST!! - 4090: Generate 20s at 720p in 2 minutes - A4500 (3070): Generate 10s at 480p in 3 minutes 1-click install on Pinokio, do it now!

LTX-2 is THE open source video AI moment I've been waiting for--the first to generates high quality video AND is FAST!! - 4090: Generate 20s at 720p in 2 minutes - A4500 (3070): Generate 10s at 480p in 3 minutes 1-click install on Pinokio, do it now!

cocktail peanut

116,072 次观看 • 7 个月前

Introducing stBTC: the Yield-Bearing Bitcoin Standard. For the first time, you can put your Bitcoin to work and earn raw BTC yield: no tokens, no inflation, no lockups. Bitcoin In → More Bitcoin Out. Cold, Hard Bitcoin Yield. In T-24 hours.

Introducing stBTC: the Yield-Bearing Bitcoin Standard. For the first time, you can put your Bitcoin to work and earn raw BTC yield: no tokens, no inflation, no lockups. Bitcoin In → More Bitcoin Out. Cold, Hard Bitcoin Yield. In T-24 hours.

Botanix 🕷️

198,874 次观看 • 11 个月前

🚨 Introducing, for the first time ever, TDM at Twitch Rivals! 🚨 Watch the The Drop In, live now, as players compete for a $200k prize pool and you have a chance at scoring a delivery code! Type #Apex4nniversary + #RespawnGiveaway in chat to enter. 👉

🚨 Introducing, for the first time ever, TDM at Twitch Rivals! 🚨 Watch the The Drop In, live now, as players compete for a $200k prize pool and you have a chance at scoring a delivery code! Type #Apex4nniversary + #RespawnGiveaway in chat to enter. 👉

Apex Legends

716,514 次观看 • 3 年前

Introducing The network for real estate intelligence & finance. First, we built the world's most accurate property valuation model. Next, we open Real Estate lending to everyone. From login to liquid in minutes.

Introducing The network for real estate intelligence & finance. First, we built the world's most accurate property valuation model. Next, we open Real Estate lending to everyone. From login to liquid in minutes.

Zipcode

50,132 次观看 • 3 个月前

Wait for it. Fast growing fires in Colorado! 120,000 acres burning with 85% of the state in severe or worse #drought. The largest is the Aspen Acres fire which has ballooned to 50K acres and 0% containment. Smoke 💨 filled skies and poor air quality expected for days. #fire #smoke #colorado #aspen #burn

Wait for it. Fast growing fires in Colorado! 120,000 acres burning with 85% of the state in severe or worse #drought. The largest is the Aspen Acres fire which has ballooned to 50K acres and 0% containment. Smoke 💨 filled skies and poor air quality expected for days. #fire #smoke #colorado #aspen #burn

Jeff Berardelli

12,268 次观看 • 2 个月前

world models have their comfyui now. introducing alakazam studio, the first local, open source UI for world models. •⁠ ⁠sentence goes in => you edit the state graph => play •⁠ ⁠on a live model, yours, running in your browser •⁠ ⁠⁠apache-2.0, fully open source fork it and point it at your model 👇

world models have their comfyui now. introducing alakazam studio, the first local, open source UI for world models. •⁠ ⁠sentence goes in => you edit the state graph => play •⁠ ⁠on a live model, yours, running in your browser •⁠ ⁠⁠apache-2.0, fully open source fork it and point it at your model 👇

hugo

25,362 次观看 • 20 天前

GROK 4.1 FAST JUST DID THE IMPOSSIBLE AGAIN: 334 BILLION TOKENS IN A SINGLE DAY The OpenRouter leaderboard isn’t just broken. It’s been nuked from orbit. Yesterday: 253B tokens Today: 334B tokens That’s an 81B token jump in 24 hours. The closest competitor needs almost an entire week to hit what Grok 4.1 Fast does before lunch. This is the free model. The one anyone can use with zero paywall. People aren’t testing Grok anymore. They’ve moved in permanently. The throne isn’t up for grabs. It’s already been welded shut with Grok sitting on it. Bow down. The king just went super Saiyan. Source: X Freeze

GROK 4.1 FAST JUST DID THE IMPOSSIBLE AGAIN: 334 BILLION TOKENS IN A SINGLE DAY The OpenRouter leaderboard isn’t just broken. It’s been nuked from orbit. Yesterday: 253B tokens Today: 334B tokens That’s an 81B token jump in 24 hours. The closest competitor needs almost an entire week to hit what Grok 4.1 Fast does before lunch. This is the free model. The one anyone can use with zero paywall. People aren’t testing Grok anymore. They’ve moved in permanently. The throne isn’t up for grabs. It’s already been welded shut with Grok sitting on it. Bow down. The king just went super Saiyan. Source: X Freeze

Mario Nawfal

42,979 次观看 • 9 个月前

🚨 SplatTransform 2.0 is here!🚨 Open source CLI tool and library for 3DGS processing NEW: Fully automated, lightning fast creation of high quality collision for your splats. Outputs: 🧊 Voxel format 🔼 Mesh format See both formats in action in the video. Links 👇

🚨 SplatTransform 2.0 is here!🚨 Open source CLI tool and library for 3DGS processing NEW: Fully automated, lightning fast creation of high quality collision for your splats. Outputs: 🧊 Voxel format 🔼 Mesh format See both formats in action in the video. Links 👇

PlayCanvas

35,795 次观看 • 3 个月前

PAYING PER MODEL IS THE DUMBEST THING IN TECH RIGHT NOW i was paying 3x what i needed to for AI inference the grid lets you buy a quality spec instead of a specific model.. it routes every request in real time to the cheapest option that qualifies swap one url and your code keeps working exactly the same openai-compatible, one line to switch, 200M free tokens to start

PAYING PER MODEL IS THE DUMBEST THING IN TECH RIGHT NOW i was paying 3x what i needed to for AI inference the grid lets you buy a quality spec instead of a specific model.. it routes every request in real time to the cheapest option that qualifies swap one url and your code keeps working exactly the same openai-compatible, one line to switch, 200M free tokens to start

Robin Delta

15,729 次观看 • 3 个月前