In June, the NinjaTech team reached out to Cerebras with the ambitious goal of cutting deep research time down to 1min at frontier quality. We reached the speed, but the models at the time weren’t quite smart enough vs. o3.
LLMS
-
SuperNinja: 5x Faster Deep Research AI Powered by Cerebras
By
–
Introducing SuperNinja – the world’s fastest deep research by @NinjaTechAI powered by Cerebras:
— Cerebras (@cerebras) 15 août 2025
‣ 1-2min for full answer instead of 10min
‣ 5x faster than o3, Gemini Pro etc.
‣ Same accuracy as measured by GAIA pic.twitter.com/rXDoamiksmIntroducing SuperNinja – the world’s fastest deep research by @NinjaTechAI powered by Cerebras:
‣ 1-2min for full answer instead of 10min
‣ 5x faster than o3, Gemini Pro etc.
‣ Same accuracy as measured by GAIA -
Character-Level Tokenization Success in DNA Language Models
By
–
watch me realize in realtime that @gdb is deadly serious about character-level tokenization and it's already working well for @pdhsu, @SKonermann, and @patrickc's @ArcInstitute 's 40B model that is the "GPT-1 of DNA"
— swyx 🐣 (@swyx) 15 août 2025
with biological language models (instead of natural language),… https://t.co/SMA81JYX0b pic.twitter.com/XXsMmDEZWewatch me realize in realtime that @gdb is deadly serious about character-level tokenization and it's already working well for @pdhsu
, @SKonermann
, and @patrickc
's @ArcInstitute 's 40B model that is the "GPT-1 of DNA" with biological language models (instead of natural language), -
LangGraph Studio Trace Mode: Real-time LangSmith Integration
By
–
🔍New in LangGraph Studio: Trace mode
— LangChain (@LangChain) 15 août 2025
View your LangSmith traces in real time right inside Studio.
Annotate runs, and add them to datasets or annotation queues, bringing the power of LangSmith tracing directly into your workflow.
Debug faster and dig deeper, without any of the… pic.twitter.com/jt5mFDDtXGNew in LangGraph Studio: Trace mode
View your LangSmith traces in real time right inside Studio.
Annotate runs, and add them to datasets or annotation queues, bringing the power of LangSmith tracing directly into your workflow.
Debug faster and dig deeper, without any of the -
Denny Zhou’s Pioneering Contributions to Modern LLM Reasoning
By
–
Many aspects of modern LLM reasoning were invented by @denny_zhou and his team. Just putting it out there because many people do not know what goes on in frontier labs.
-
Warmth and Factuality: An Inverse Trade-off
By
–
Optimizing for increased warmth produces decreased factuality.
-
GPT-5 Pricing and Performance Compared to Mini and Chat Models
By
–
That's very interesting – I thought 'mini' might rank a bit higher than this. 'Chat' being below GPT-4o is also kind of interesting, considering it is similarly priced (GPT-5 input is cheaper, but output same)
-
CulturalGround Open-Source Multilingual Cultural VQA Dataset Released
By
–
The code used for curating CulturalGround, the largest open-source multilingual cultural VQA dataset, is now publicly available. The proposed pipeline could greatly enhance factual understanding of a wide array of real-world entities in multimodal LLMs, at zero cost. 30M VQA
-
Dense Models Outperform Large Models for Practical Applications
By
–
1-20b doesn't benefit almost any normal person situation. Whereas an equivalent 4b dense is usable almost anywhere. Qwen3 Coder doesn't even have any dense option.