Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

open sourcing Marlin-2B 🐟 a tiny VLM to extract structured information from videos Marlin is finetuned for two questions devs want to ask in their videos: what is happening, and when? Best open model in its weight class, competitive with Gemini-2.5-flash at only 2B params 🧵

Shubham Sharma

1,215 subscribers

307,394 views • 1 month ago •via X (Twitter)

Science & Technology

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Introducing HYPERWORM: The first verifiable digital organism onchain. Our open-source model adapted from Nematoduino runs in Marlin TEE network on Hyperliquid EVM. Visit: 🧵🪱

Introducing HYPERWORM: The first verifiable digital organism onchain. Our open-source model adapted from Nematoduino runs in Marlin TEE network on Hyperliquid EVM. Visit: 🧵🪱

Project Nema

195,551 views • 1 year ago

The plot thickens. 🕵️ Use Gemini 2.5 Pro in the Google Gemini to turn random videos from your camera roll into a dramatic narrative. Try it yourself: 1️⃣ Go to Gemini and select “2.5 Pro” from the model dropdown. 2️⃣ Tap the plus icon and upload three videos from your camera roll. (Pro/Ultra subscribers, try using your Veo creations! 🎥) 3️⃣ Ask Gemini to write a short story based on the videos. Here’s our prompt to get you started: “Come up with a premise for an epic adventure based on these videos.”

The plot thickens. 🕵️ Use Gemini 2.5 Pro in the Google Gemini to turn random videos from your camera roll into a dramatic narrative. Try it yourself: 1️⃣ Go to Gemini and select “2.5 Pro” from the model dropdown. 2️⃣ Tap the plus icon and upload three videos from your camera roll. (Pro/Ultra subscribers, try using your Veo creations! 🎥) 3️⃣ Ask Gemini to write a short story based on the videos. Here’s our prompt to get you started: “Come up with a premise for an epic adventure based on these videos.”

Google

164,396 views • 1 year ago

Announcing Talkie: a new, open-weight historical LLM! We trained and finetuned a 13B model on a newly-curated dataset of only pre-1930 data. Try it below! with Alec Radford and Nick Levine 🧵

Announcing Talkie: a new, open-weight historical LLM! We trained and finetuned a 13B model on a newly-curated dataset of only pre-1930 data. Try it below! with Alec Radford and Nick Levine 🧵

David Duvenaud

1,424,500 views • 1 month ago

Wike Is Still Open To Negotiation Fubara and Wike should engage in dialogue instead of creating problems for Ijaw land. Pioneer Secretary, APC Bayelsa State, Marlin Daniel

Wike Is Still Open To Negotiation Fubara and Wike should engage in dialogue instead of creating problems for Ijaw land. Pioneer Secretary, APC Bayelsa State, Marlin Daniel

ARISE NEWS

50,328 views • 1 year ago

Gemini 2.5 Flash Preview is an amazing model. Google is literally winning. No stopping them now, this is not normal. Gemini 2.5 Flash is very cheap and super intelligent model. Intelligence too cheap to meter this is what it means. Thank you, Google.

Gemini 2.5 Flash Preview is an amazing model. Google is literally winning. No stopping them now, this is not normal. Gemini 2.5 Flash is very cheap and super intelligent model. Intelligence too cheap to meter this is what it means. Thank you, Google.

AshutoshShrivastava

154,366 views • 1 year ago

Michael Jordan is competing in the White Marlin Open this weekend with his 82-foot Bayliss fishing boat. MJ and the Catch 23 crew are currently in second place after weighing in a 71-pound white marlin. If the result holds, they'll walk away with $400,000 in prize money.

Michael Jordan is competing in the White Marlin Open this weekend with his 82-foot Bayliss fishing boat. MJ and the Catch 23 crew are currently in second place after weighing in a 71-pound white marlin. If the result holds, they'll walk away with $400,000 in prize money.

Joe Pompliano

4,775,299 views • 10 months ago

ACE-Step-v1.5(2B) is a new open-source music AI model that runs locally in #ComfyUI :) * Only needs ~4GB VRAM * MIT licence * Done in 8 seconds! * Have a listen :p

ACE-Step-v1.5(2B) is a new open-source music AI model that runs locally in #ComfyUI :) * Only needs ~4GB VRAM * MIT licence * Done in 8 seconds! * Have a listen :p

Nerdy Rodent 🐀🤓💻🪐🚴

11,663 views • 4 months ago

This is wild. In just a few hours my friend forked Open AI Codex to work with Gemini 2.5 Pro 🤯

This is wild. In just a few hours my friend forked Open AI Codex to work with Gemini 2.5 Pro 🤯

Sawyer Hood

314,120 views • 1 year ago

Heaven Birthday, "Marlin the Magician"🕯️ Marlin Briscoe In 1968 while with the #Broncos, Marlin became the first Black player to start at quarterback in the Super Bowl era. After being denied an opportunity to compete for the starting QB job in '69, he moved on to the #Bills, where he would earn Pro Bowl honors as a receiver in 1970. Marlin later caught on with the #Dolphins, becoming a solid role player as a receiver on their 1972 and '73 Super Bowl championship teams. He retired as a player in 1977 after having brief stints with the Chargers, Lions, and Patriots.

Heaven Birthday, "Marlin the Magician"🕯️ Marlin Briscoe In 1968 while with the #Broncos, Marlin became the first Black player to start at quarterback in the Super Bowl era. After being denied an opportunity to compete for the starting QB job in '69, he moved on to the #Bills, where he would earn Pro Bowl honors as a receiver in 1970. Marlin later caught on with the #Dolphins, becoming a solid role player as a receiver on their 1972 and '73 Super Bowl championship teams. He retired as a player in 1977 after having brief stints with the Chargers, Lions, and Patriots.

Kevin Gallagher

20,540 views • 9 months ago

Excited to announce GR00T N1, the world’s first open foundation model for humanoid robots! We are on a mission to democratize Physical AI. The power of general robot brain, in the palm of your hand - with only 2B parameters, N1 learns from the most diverse physical action dataset ever compiled and punches above its weight: - Real humanoid teleoperation data. - Large-scale simulation data: we are open-sourcing 300K+ trajectories! - Neural trajectories: we apply SOTA video generation models to “hallucinate” new synthetic data that features accurate physics in pixels. Using Jensen’s words, “systematically infinite data”! - Latent actions: we develop novel algorithms to extract action tokens from in-the-wild human videos and neural generated videos. GR00T N1 is a single end-to-end neural net, from photons to actions: - Vision-Language Model (System 2) that interprets the physical world through vision and language instructions, enabling robots to reason about their environment and instructions, and plan the right actions. - Diffusion Transformer (System 1) that “renders” smooth and precise motor actions at 120 Hz, executing the latent plan made by System 2. We deploy N1 on GR1 robot, 1X Neo robot, and a large collection of simulation benchmarks. N1 achieves up to +30% boost in diverse manipulation tasks for household and industrial settings. While humanoid robots are the main focus of N1, our model also supports cross-embodiment. We finetune it to work on the $110 HuggingFace LeRobot SO100 robot arm! Open robot brain runs on open hardware. Sounds just right. Let’s solve robotics, together, one token at a time. Links to our Whitepaper, Github repo, HuggingFace model, and open dataset page in the thread: 🧵

Excited to announce GR00T N1, the world’s first open foundation model for humanoid robots! We are on a mission to democratize Physical AI. The power of general robot brain, in the palm of your hand - with only 2B parameters, N1 learns from the most diverse physical action dataset ever compiled and punches above its weight: - Real humanoid teleoperation data. - Large-scale simulation data: we are open-sourcing 300K+ trajectories! - Neural trajectories: we apply SOTA video generation models to “hallucinate” new synthetic data that features accurate physics in pixels. Using Jensen’s words, “systematically infinite data”! - Latent actions: we develop novel algorithms to extract action tokens from in-the-wild human videos and neural generated videos. GR00T N1 is a single end-to-end neural net, from photons to actions: - Vision-Language Model (System 2) that interprets the physical world through vision and language instructions, enabling robots to reason about their environment and instructions, and plan the right actions. - Diffusion Transformer (System 1) that “renders” smooth and precise motor actions at 120 Hz, executing the latent plan made by System 2. We deploy N1 on GR1 robot, 1X Neo robot, and a large collection of simulation benchmarks. N1 achieves up to +30% boost in diverse manipulation tasks for household and industrial settings. While humanoid robots are the main focus of N1, our model also supports cross-embodiment. We finetune it to work on the $110 HuggingFace LeRobot SO100 robot arm! Open robot brain runs on open hardware. Sounds just right. Let’s solve robotics, together, one token at a time. Links to our Whitepaper, Github repo, HuggingFace model, and open dataset page in the thread: 🧵

Jim Fan

465,769 views • 1 year ago

Thanks to the pro-growth tax reforms of H.R. 1, Marlin Steel is making its largest investment in company history. #ManufacturingWins Drew Greenblatt Marlin Steel Wire Griff Jenkins FOX & Friends

Thanks to the pro-growth tax reforms of H.R. 1, Marlin Steel is making its largest investment in company history. #ManufacturingWins Drew Greenblatt Marlin Steel Wire Griff Jenkins FOX & Friends

The NAM

55,436 views • 5 months ago

Inferact cofounder Simon Mo: “I fundamentally believe that open source, especially how vLLM itself is structured, is critical to the AI infrastructure in the world. And what we want to do with Inferact is to support, maintain, steward and push forward the open source ecosystem.” “Open source moves so fast that the only way to stay ahead is adopting... And in fact this is exactly why we're staying all in on open source.” Simon Mo

Inferact cofounder Simon Mo: “I fundamentally believe that open source, especially how vLLM itself is structured, is critical to the AI infrastructure in the world. And what we want to do with Inferact is to support, maintain, steward and push forward the open source ecosystem.” “Open source moves so fast that the only way to stay ahead is adopting... And in fact this is exactly why we're staying all in on open source.” Simon Mo

a16z

29,383 views • 5 months ago

NEW: OpenAI releases Privacy Filter, their first open model of 2026! 🤗 Apache-2.0! It's a bidirectional token-classification adaptation of GPT-OSS, trained to mask personally identifiable information (PII) in text. At only 1.5B params, it can even run locally in your browser!

NEW: OpenAI releases Privacy Filter, their first open model of 2026! 🤗 Apache-2.0! It's a bidirectional token-classification adaptation of GPT-OSS, trained to mask personally identifiable information (PII) in text. At only 1.5B params, it can even run locally in your browser!

Xenova

219,738 views • 2 months ago

"The Tower Heist" A small window is open to recover an ancient scroll. Created in early access with the fantastic new Kling AI 2.5 Turbo model bringing affordable generations with best in class prompt adherence and quality. The range of directed character motions now available to us is incredible! 🔊🔊🎧

"The Tower Heist" A small window is open to recover an ancient scroll. Created in early access with the fantastic new Kling AI 2.5 Turbo model bringing affordable generations with best in class prompt adherence and quality. The range of directed character motions now available to us is incredible! 🔊🔊🎧

Stevie Mac

114,555 views • 9 months ago

Fast matters. LTX-2 is the fastest open-source model in its class, up to 18x faster than competitors.

Fast matters. LTX-2 is the fastest open-source model in its class, up to 18x faster than competitors.

LTX

18,877 views • 5 months ago

Google's best model is actually hidden Gemini 2.5 Pro with Web Grounding is incredible and only available in AI Studio You get free access to the most powerful model with real-time data: - Access AI Studio - Select Gemini 2.5 Pro Preview - Activate Grounding with Search Gemini will search the data, think, then give you the answers with the sources.

Google's best model is actually hidden Gemini 2.5 Pro with Web Grounding is incredible and only available in AI Studio You get free access to the most powerful model with real-time data: - Access AI Studio - Select Gemini 2.5 Pro Preview - Activate Grounding with Search Gemini will search the data, think, then give you the answers with the sources.

Paul Couvert

149,669 views • 1 year ago

ELON: CHINA HAS THE BEST OPEN SOURCE AI MODELS RIGHT NOW “The best open source models are generally from China, which is bizarre. I think the second best one, or maybe it’s better than second best, is Grok 2.5. The open source model is actually very good, and we will continue to open source our models.” Source: The All-In Podcast

ELON: CHINA HAS THE BEST OPEN SOURCE AI MODELS RIGHT NOW “The best open source models are generally from China, which is bizarre. I think the second best one, or maybe it’s better than second best, is Grok 2.5. The open source model is actually very good, and we will continue to open source our models.” Source: The All-In Podcast

Mario Nawfal

1,163,582 views • 7 months ago

#New Rep. Marlin Stutzman says Secretary Marco Rubio is “very concerned” about the situation in Middle East [Syria]. Rep. Marlin Stutzman told me - Rudaw English that when there are attacks on Kurds or Druze, “the responsibility goes back on President al-Sharaa” to get it under control, and to ask for help if needed.

#New Rep. Marlin Stutzman says Secretary Marco Rubio is “very concerned” about the situation in Middle East [Syria]. Rep. Marlin Stutzman told me - Rudaw English that when there are attacks on Kurds or Druze, “the responsibility goes back on President al-Sharaa” to get it under control, and to ask for help if needed.

Diyar Kurda

11,726 views • 4 months ago

Pentagon in chaos! Pete Hegseth is at the center of controversies and Marlin Stutzman for Congress came on TMZ Live to defend the madness.

Pentagon in chaos! Pete Hegseth is at the center of controversies and Marlin Stutzman for Congress came on TMZ Live to defend the madness.

TMZ Live

27,518 views • 1 year ago