Hugging Face (Twitter)
RT @skalskip92: this is so much fun! https://twitter.com/skalskip92/status/2002124700647047460#m
Post #2074
24
Channel Public Channel
HU Hugging Face
@huggingface
- Subscribers
- 302
- Photos
- 1.3K
- Videos
- 408
- Links
- 2.2K
Showing posts older than #2075 · Back to latest
Older Posts 20 shown
Post #2073
22
Hugging Face (Twitter)
RT @Tu7uruu: Took some days off and finally had time to read the ultrascla playbook
A bit late but it’s really goated
https://huggingface.co/spaces/nanotron/ultrascale-playbook?section=high-level_overview
RT @Tu7uruu: Took some days off and finally had time to read the ultrascla playbook
A bit late but it’s really goated
https://huggingface.co/spaces/nanotron/ultrascale-playbook?section=high-level_overview
Post #2072
29
Hugging Face (Twitter)
RT @victormustar: my 128GB macbook is getting smart 🔥
`llama-server -hf unsloth/MiniMax-M2.1-GGUF:Q2_K_XL`
(2026 is the year of local agents)
RT @victormustar: my 128GB macbook is getting smart 🔥
`llama-server -hf unsloth/MiniMax-M2.1-GGUF:Q2_K_XL`
(2026 is the year of local agents)
Post #2071
31
Hugging Face (Twitter)
RT @victormustar: MinMax shared some gold on Hugging Face (deep dive on agent generalization & alignment)
https://huggingface.co/blog/MiniMax-AI/aligning-to-what
RT @victormustar: MinMax shared some gold on Hugging Face (deep dive on agent generalization & alignment)
https://huggingface.co/blog/MiniMax-AI/aligning-to-what
Post #2070
23
Hugging Face (Twitter)
RT @novita_labs: 🤗 MiniMax-M2.1 is now live on @huggingface , supported by Novita
A 10B-activated open-source coding & agent model from MiniMax, with strong performance on SWE-multilingual and VIBE-bench benchmarks.
Test it yourself 👇
RT @novita_labs: 🤗 MiniMax-M2.1 is now live on @huggingface , supported by Novita
A 10B-activated open-source coding & agent model from MiniMax, with strong performance on SWE-multilingual and VIBE-bench benchmarks.
Test it yourself 👇
Post #2069
24
Hugging Face (Twitter)
RT @Michael_D_Moor: What a marvelous christmas present by @huggingface 🤖🚂
#reachymini
RT @Michael_D_Moor: What a marvelous christmas present by @huggingface 🤖🚂
#reachymini
Post #2068
29
Hugging Face (Twitter)
RT @kimmonismus: This is nuts: 42% in GPQA for an 2.6B model!
This can run on your iPhone and has phd knowledge. Insane https://twitter.com/liquidai/status/2004190178068296181#m
RT @kimmonismus: This is nuts: 42% in GPQA for an 2.6B model!
This can run on your iPhone and has phd knowledge. Insane https://twitter.com/liquidai/status/2004190178068296181#m
Post #2067
28
Hugging Face (Twitter)
RT @ArturSkowronski: How’s Christmas Day going for you? My daughter and I spent it assembling Reachy Mini ❤️
It wasn’t planned as a Christmas gift, but thanks to customs checks, courier delays, and the fact that @huggingface and @pollenrobotics themselves slipped production by a quarter (no hard feelings), it finally reached (pun intended) me on December 23rd - so it ended up under the Christmas tree 😃
It took us three hours to assemble, and it was freaking fun for both of us... though mostly for me, I think 😆
RT @ArturSkowronski: How’s Christmas Day going for you? My daughter and I spent it assembling Reachy Mini ❤️
It wasn’t planned as a Christmas gift, but thanks to customs checks, courier delays, and the fact that @huggingface and @pollenrobotics themselves slipped production by a quarter (no hard feelings), it finally reached (pun intended) me on December 23rd - so it ended up under the Christmas tree 😃
It took us three hours to assemble, and it was freaking fun for both of us... though mostly for me, I think 😆
Post #2066
27
Hugging Face (Twitter)
RT @liquidai: Meet the strongest 3B model on the market.
LFM2-2.6B-Exp is an experimental checkpoint built on LFM2-2.6B using pure reinforcement learning.
> Consistent improvements in instruction following, knowledge, and math benchmarks
> Outperforms other 3B models in these domains
> Its IFBench score surpasses DeepSeek R1-0528, a model 263x larger
Download and play 👉 https://huggingface.co/LiquidAI/LFM2-2.6B-Exp
Happy holidays,
The Liquid AI team🎄✨
RT @liquidai: Meet the strongest 3B model on the market.
LFM2-2.6B-Exp is an experimental checkpoint built on LFM2-2.6B using pure reinforcement learning.
> Consistent improvements in instruction following, knowledge, and math benchmarks
> Outperforms other 3B models in these domains
> Its IFBench score surpasses DeepSeek R1-0528, a model 263x larger
Download and play 👉 https://huggingface.co/LiquidAI/LFM2-2.6B-Exp
Happy holidays,
The Liquid AI team🎄✨
Post #2065
22
Hugging Face (Twitter)
RT @zRdianjiao: Been busy, still have many Pull Requests to submit. After Christmas, it will be a busy review process https://twitter.com/Zai_org/status/2003828175089098943#m
RT @zRdianjiao: Been busy, still have many Pull Requests to submit. After Christmas, it will be a busy review process https://twitter.com/Zai_org/status/2003828175089098943#m
Post #2064
32
Hugging Face (Twitter)
RT @dkundel: I was going to take a break from coding over the holidays...I guess now I have to ask Codex to tap in and help get my @huggingface Reachy Mini up and running 😄
RT @dkundel: I was going to take a break from coding over the holidays...I guess now I have to ask Codex to tap in and help get my @huggingface Reachy Mini up and running 😄
Post #2063
28
Hugging Face (Twitter)
RT @ehcalabres: 🎁 Last-minute gift from Santa! 🎅
FOMO over the Reachy Mini by @pollenrobotics & @huggingface? No hardware, no problem!
Build and prototype your AI apps for free in the simulation. The only catch? You can’t pet him! 🤖
Just install the Python SDK & start playing!
RT @ehcalabres: 🎁 Last-minute gift from Santa! 🎅
FOMO over the Reachy Mini by @pollenrobotics & @huggingface? No hardware, no problem!
Build and prototype your AI apps for free in the simulation. The only catch? You can’t pet him! 🤖
Just install the Python SDK & start playing!
Post #2062
22
Hugging Face (Twitter)
RT @Zai_org: Trending #1 on @huggingface again! We’ll keep open-sourcing next year.
RT @Zai_org: Trending #1 on @huggingface again! We’ll keep open-sourcing next year.
Post #2061
27
Hugging Face (Twitter)
RT @NielsRogge: You might have seen the very cool video of @victormustar showing hf.co/chat calling the Z-Image-Turbo model by @Ali_TongyiLab via MCP
How to do this?
1. Go to https://huggingface.co/spaces/Tongyi-MAI/Z-Image-Turbo
2. Click the little "MCP" icon at the top
3. Click "add to MCP tools"
4. Profit
RT @NielsRogge: You might have seen the very cool video of @victormustar showing hf.co/chat calling the Z-Image-Turbo model by @Ali_TongyiLab via MCP
How to do this?
1. Go to https://huggingface.co/spaces/Tongyi-MAI/Z-Image-Turbo
2. Click the little "MCP" icon at the top
3. Click "add to MCP tools"
4. Profit
Post #2060
23
Post #2059
24
Hugging Face (Twitter)
RT @LeRobotHF: We’ve teamed up with @XSquareRobot to integrate WALL-OSS, a powerful new VLA foundation model into LeRobot!
RT @LeRobotHF: We’ve teamed up with @XSquareRobot to integrate WALL-OSS, a powerful new VLA foundation model into LeRobot!
Post #2058
29
Hugging Face (Twitter)
RT @pa_balland: Something big is happening in robotics - and it’s hiding in plain sight.
This post is not about dancing robots but in the data that powers them. Open robotics datasets have exploded this year, turning the field into a more scalable and collaborative ecosystem.
In just two years, @huggingface datasets grew from 11k to over 600k - and robotics is by far the fastest-growing segment. We went from 1k robotics datasets in 2024 to 27k in 2025!
For comparison, text generation, the second-largest category, has only around 5k datasets in 2025. That gap is massive.
Open datasets are important because robotics lives and dies by real-world robot data - video, actions, sensors, failures. By making this data easy to upload, reuse, and benchmark, researchers, startups, and large players are now releasing real-robot datasets that would have stayed locked inside labs just a few years ago.
Major contributors include @nvidia, LeRobot initiative, and...
Перейти на оригинальный пост
RT @pa_balland: Something big is happening in robotics - and it’s hiding in plain sight.
This post is not about dancing robots but in the data that powers them. Open robotics datasets have exploded this year, turning the field into a more scalable and collaborative ecosystem.
In just two years, @huggingface datasets grew from 11k to over 600k - and robotics is by far the fastest-growing segment. We went from 1k robotics datasets in 2024 to 27k in 2025!
For comparison, text generation, the second-largest category, has only around 5k datasets in 2025. That gap is massive.
Open datasets are important because robotics lives and dies by real-world robot data - video, actions, sensors, failures. By making this data easy to upload, reuse, and benchmark, researchers, startups, and large players are now releasing real-robot datasets that would have stayed locked inside labs just a few years ago.
Major contributors include @nvidia, LeRobot initiative, and...
Перейти на оригинальный пост
- ❤ 1
Post #2057
31
Hugging Face (Twitter)
RT @Zai_org: GLM-4.7 is here!
GLM-4.7 surpasses GLM-4.6 with substantial improvements in coding, complex reasoning, and tool usage, setting new open-source SOTA standards. It also boosts performance in chat, creative writing, and role-play scenarios.
Default Model for Coding Plan: z.ai/subscribe
Try it now: chat.z.ai
Weights: http://huggingface.co/zai-org/GLM-4.7
Tech Blog: z.ai/blog/glm-4.7
RT @Zai_org: GLM-4.7 is here!
GLM-4.7 surpasses GLM-4.6 with substantial improvements in coding, complex reasoning, and tool usage, setting new open-source SOTA standards. It also boosts performance in chat, creative writing, and role-play scenarios.
Default Model for Coding Plan: z.ai/subscribe
Try it now: chat.z.ai
Weights: http://huggingface.co/zai-org/GLM-4.7
Tech Blog: z.ai/blog/glm-4.7
Post #2056
26
Post #2055
26
