15 91 387

alkinun

AtAndDev

AI & ML interests

LLMs, Alignment, Merging, Unsloth, DPO, SFT, ORPO, SPIN..

Recent Activity

liked a model 1 day ago

facebook/KernelLLM

reacted to Jofthomas's post with 👍 1 day ago

Meet our new agentic model : 𝗗𝗲𝘃𝘀𝘁𝗿𝗮𝗹 Devstral is an open-source LLM built software engineering tasks built under a collaboration between Mistral AI and All Hands AI 🙌. 𝗞𝗲𝘆 𝗳𝗲𝗮𝘁𝘂𝗿𝗲𝘀 : • 🤖 𝗔𝗴𝗲𝗻𝘁𝘀 : perfect for Agentic coding • 🍃 𝗹𝗶𝗴𝗵𝘁𝘄𝗲𝗶𝗴𝗵𝘁: Devstral is a 𝟮𝟰𝗕 parameter based on Mistral small. • ©️ 𝗔𝗽𝗮𝗰𝗵𝗲 𝟮.𝟬, meaning fully open-source ! • 📄 A 𝟭𝟮𝟴𝗸 context window. 📚Blog : https://mistral.ai/news/devstral ⚡API : The model is also available on our API under the name 𝗱𝗲𝘃𝘀𝘁𝗿𝗮𝗹-𝘀𝗺𝗮𝗹𝗹-𝟮𝟱𝟬𝟱 🤗 repo : https://huggingface.co/mistralai/Devstral-Small-2505 Can't wait to see what you will build with it !

reacted to Jofthomas's post with 🔥 1 day ago

View all activity

Organizations

AtAndDev's activity

reacted to Jofthomas's post with 👍🔥 1 day ago

Post

2222

Meet our new agentic model : 𝗗𝗲𝘃𝘀𝘁𝗿𝗮𝗹

Devstral is an open-source LLM built software engineering tasks built under a collaboration between Mistral AI and All Hands AI 🙌.

𝗞𝗲𝘆 𝗳𝗲𝗮𝘁𝘂𝗿𝗲𝘀 :
• 🤖 𝗔𝗴𝗲𝗻𝘁𝘀 : perfect for Agentic coding
• 🍃 𝗹𝗶𝗴𝗵𝘁𝘄𝗲𝗶𝗴𝗵𝘁: Devstral is a 𝟮𝟰𝗕 parameter based on Mistral small.
• ©️ 𝗔𝗽𝗮𝗰𝗵𝗲 𝟮.𝟬, meaning fully open-source !
• 📄 A 𝟭𝟮𝟴𝗸 context window.

📚Blog : https://mistral.ai/news/devstral
⚡API : The model is also available on our API under the name 𝗱𝗲𝘃𝘀𝘁𝗿𝗮𝗹-𝘀𝗺𝗮𝗹𝗹-𝟮𝟱𝟬𝟱
🤗 repo : mistralai/Devstral-Small-2505

Can't wait to see what you will build with it !

1 reply

replied to merve's post 1 day ago

Such a cool implementation of a model
Btw ben de türküm :)

reacted to merve's post with 🚀 1 day ago

Post

2676

Bu post'u çevirebilirsiniz 🤗💗

6 replies

reacted to YerbaPage's post with 🤗👀🔥 5 days ago

Post

2971

Curated list of **Next Gen Code Generation** papers & benchmarks! 🔥 with 50+ ⭐️ now!

Stay ahead with the latest in:
✅ Repo-level Issue Resolution (SWE-bench, Agents)
✅ Repo-level Code Completion (Repo understanding)
✅ Datasets & Benchmarks

👉 Check it out: https://github.com/YerbaPage/Awesome-Repo-Level-Code-Generation 🔥
💡PRs are welcomed!

reacted to regisss's post with 🚀 7 days ago

Post

2314

It will be very interesting to benchmark how energy-efficient the new Falcon-Edge is: https://huggingface.co/blog/tiiuae/falcon-edge

Should be super efficient on CPU according to the numbers published for BitNet: https://github.com/microsoft/BitNet/blob/main/assets/intel_performance.jpg

1 reply

replied to ProCreations's post 9 days ago

Well, a gpu (like 10k) has MUCH MUCH more cores than a cpu (like 12) and cpus have much more capable cores. GPUs just do simple matrix mul. so we would have the same num of cores with gpus but the cores will become much more capable. Which is a big W. Sorry for the nerdieness but I just had to hop in. :)

reacted to hesamation's post with 👍🔥❤️ about 1 month ago

Post

2914

this paper has been blowing up
they train an open-source multimodal LLM (InternVL3) that can compete with GPT-4o and Claude 3.5 Sonnet by:
> training text and vision on a single stage
> a novel V2PE positional encoding
> SFT & mixed preference optimization
Paper: InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models (2504.10479)
> test-time scaling

replied to Steven10429's post about 1 month ago

It seems like they randomly filter out some ppl for some reason...

reacted to AdinaY's post with 👍 about 1 month ago

Post

1839

MAYE🎈a from-scratch RL framework for Vision Language Models, released by GAIR - an active research group from the Chinese community.

✨Minimal & transparent pipeline with standard tools
✨Standardized eval to track training & reflection
✨Open Code & Dataset

Code:
https://github.com/GAIR-NLP/MAYE?tab=readme-ov-file
Dataset:
ManTle/MAYE
Paper:
Rethinking RL Scaling for Vision Language Models: A Transparent, From-Scratch Framework and Comprehensive Evaluation Scheme (2504.02587)

1 reply

reacted to merterbak's post with ❤️👀🔥 about 1 month ago

Post

4755

Qwen 3 can launch very soon. 👀

https://github.com/ggml-org/llama.cpp/pull/12828

3 replies

replied to their post about 2 months ago

No one has

posted an update about 2 months ago

Post

2987

Llama 4 is out...

3 replies

reacted to BestWishYsh's post with 👀 about 2 months ago

Post

2757

🚨 Hot Take: GPT-4o might NOT be a purely autoregressive model! 🚨

There’s a high chance it has a diffusion head. 🤯 If true, this could be a game-changer for AI architecture. What do you think? 🤔👇

Code: https://github.com/PicoTrex/GPT-ImgEval
Dataset: Yejy53/GPT-ImgEval
Paper: GPT-ImgEval: A Comprehensive Benchmark for Diagnosing GPT4o in Image Generation (2504.02782)