gabrielmbmb_ Profile Banner
Gabriel Martín Blázquez Profile
Gabriel Martín Blázquez

@gabrielmbmb_

Followers
678
Following
3K
Media
62
Statuses
569

Founding Engineer | ex-Hugging Face 🤗

Madrid
Joined March 2021
Don't wanna be here? Send us removal request.
@gabrielmbmb_
Gabriel Martín Blázquez
17 days
How good is GPT-5 without reasoning? GPT-5 no reasoning > GPT 4.1?.
0
0
0
@gabrielmbmb_
Gabriel Martín Blázquez
19 days
RT @alvarobartt: 🔵 @openai just released GPT OSS (20B and 120B) on the @huggingface Hub, their first open-source LLM since GPT2 in 2019!. H….
0
5
0
@grok
Grok
5 days
What do you want to know?.
235
135
1K
@gabrielmbmb_
Gabriel Martín Blázquez
19 days
RT @alvarobartt: 🦀 @huggingface just released Text Embeddings Inference (TEI) v1.8.0, bringing a mix of brand new features and the latest a….
0
4
0
@gabrielmbmb_
Gabriel Martín Blázquez
1 month
RT @reach_vb: NEW: Higgs Audio V2 from @boson_ai open, unified TTS model w/ voice cloning, beats GPT 4o mini tts and ElevenLabs v2 🔥. > Tr….
0
130
0
@gabrielmbmb_
Gabriel Martín Blázquez
1 month
LLMs are the Thermomix of cooking for React.
0
0
1
@gabrielmbmb_
Gabriel Martín Blázquez
2 months
RT @LoubnaBenAllal1: Introducing SmolLM3: a strong, smol reasoner!. > SoTA 3B model.> dual mode reasoning (think/no_think).> long context,….
0
208
0
@gabrielmbmb_
Gabriel Martín Blázquez
3 months
when oficial mcp server @datadoghq.
0
0
1
@gabrielmbmb_
Gabriel Martín Blázquez
3 months
RT @alvarobartt: 🤗 ICYMI, yesterday at @Microsoft Build, @satyanadella announced major Azure AI updates like Azure AI Foundry Agent Service….
0
1
0
@gabrielmbmb_
Gabriel Martín Blázquez
4 months
RT @__tinygrad__: Here's the worlds first AMD GPU driven over USB3. From a Mac! Linux and Windows should work too, it's just libusb. Avail….
0
704
0
@gabrielmbmb_
Gabriel Martín Blázquez
6 months
RT @eliebakouch: Gave a talk earlier today to explain 'What's new in pre-training' to my @huggingface colleagues. I'm sharing the slides he….
0
35
0
@gabrielmbmb_
Gabriel Martín Blázquez
6 months
RT @freddy_alfonso_: Building real-time WebRTC and Websocket applications is very difficult to get right in Python. Until now - Introducin….
0
103
0
@gabrielmbmb_
Gabriel Martín Blázquez
6 months
RT @anton_lozhkov: LLM Reasoning labs will be eating good today🍔. We commandeered the HF cluster for a few days and generated 1.2M reasonin….
0
97
0
@gabrielmbmb_
Gabriel Martín Blázquez
7 months
RT @LoubnaBenAllal1: We just published the second OpenR1 update with OpenR1-220k-Math, our new large-scale dataset for mathematical reasoni….
0
61
0
@gabrielmbmb_
Gabriel Martín Blázquez
7 months
RT @simonw: Today I found out about SmolLM2-135M-Instruct, a tiny LLM which quantizes down to just below 100MB. which means it can fit in….
Tweet card summary image
simonwillison.net
I just released llm-smollm2, a new plugin for LLM that bundles a quantized copy of the SmolLM2-135M-Instruct LLM inside of the Python package. This means you can now pip install …
0
61
0
@gabrielmbmb_
Gabriel Martín Blázquez
7 months
RT @LoubnaBenAllal1: The wait is over: our SmolLM2 paper is out—a detailed guide for building SOTA small LMs. While most LM papers skim ove….
0
100
0
@gabrielmbmb_
Gabriel Martín Blázquez
7 months
Link to Hugging Face paper page:
Tweet card summary image
huggingface.co
0
0
2
@gabrielmbmb_
Gabriel Martín Blázquez
7 months
SmolLM2 paper is out! We wrote a paper detailing the steps we took to train one of the best smol LM 🤏 out there: pre-training and post-training data, training ablations and some interesting findings 💡. Go check it out and don't hesitate to write your thoughts/questions in the
Tweet media one
1
12
48
@gabrielmbmb_
Gabriel Martín Blázquez
7 months
RT @lvwerra: Intense first week for Open-R1 - time for an update!. We put all the resources in one place: a summary of progress to reproduc….
0
81
0
@gabrielmbmb_
Gabriel Martín Blázquez
7 months
DeepSeek brought the 🤯 reasoning open model, and 🤗 will bring the open recipe 👨🏽‍🍳 (with your help!). The Hugging Face team is working hard to get everyone a reproduction of the training pipeline of the mighty R1. You can follow the progress on GitHub ⤵️.
1
1
16