BREAKING : Gemini 3 Deep Think gets 41% on HLE and 45.1% on ARC_AGI-2! "In testing, Gemini 3 Deep Think outperforms Gemini 3 Pro’s already impressive performance on Humanity’s Last Exam and GPQA Diamond."
OPEN SOURCE
-
Gemini 3 Pro Benchmarks Break Records
By
–
Gemini 3 Pro benchmarks are wild – Humanity’s Last Exam: 37.5%
– ARC-AGI-2: 31.1% True SOTA -
Firecrawl: Open Source AI Web Scraping Tool
By
–
Github Repo: https://
github.com/firecrawl/fire
crawl
… Check it out here: -
GPT-5.1 tops ARC AGI 2 benchmark
By
–
GPT-5.1 Thinking High from OpenAI claims a top spot on ARC AGI 2 benchmark and dethrones Grok 4. Was GPT-5.1 undehyped?
-
AI Engineering Toolkit: Open Source Resources for Developers
By
–
Link to the GitHub repo (2K+ Github Stars): https://
github.com/Sumanth077/ai-
engineering-toolkit
… Feel free to add new tools and libraries and contribute! -
100+ LLM Libraries and Frameworks for AI Engineering
By
–
AI Engineering Toolkit! I have curated a list of 100+ LLM libraries and frameworks for training, fine-tuning, building, evaluating, deploying, RAG, and AI Agents. Categories of LLM Libraries include: • Vector Databases – Store and retrieve embeddings efficiently.
• -
China’s Moonshot AI Releases Kimi K2 Thinking Model
By
–
Pendant ce temps, la Chine avance. Moonshot AI publie Kimi K2 Thinking,
un modèle open source qui rivalise avec GPT-5. Le match EU / Chine va-t-il bientôt se jouer en orbite ?
Comme une impression de déjà-vu… -
Grok 5: 6T params, twice Grok 4
By
–
Grok 5 is expected to arrive in the first quarter of next year and be twice as big as Grok 4 according to Elon Musk. 6T params
-
GPT-5.1 Scores 76.3% on SWE-bench Verified
By
–
GPT-5.1 achieved 76.3% on SWE-bench Verified. Quite a big leap! When GPT-5.1 Pro?
-
ERNIE-4.5-VL-28B-A3B-Thinking Open-Sourced Under Apache 2.0
By
–
For developers and businesses, this is a major win: ERNIE-4.5-VL-28B-A3B-Thinking is fully open-sourced under Apache 2.0, meaning it’s ready for commercial use.