βοΈ GROK JUST GREW A BRAIN: Upload β Instant Answers ππ§
Grok can now inhale PDFs, code, CSVsβwhatever docs users chuck at itβand spit out real answers. No coffee breaks, no βfile not found,β just 24/7 research slave energy. Multi-file? Easy. Context that actually makes sense? Also yes.
#AI #Grok #TechLife #NoMoreManualReading
π @missedminutes π
Grok can now inhale PDFs, code, CSVsβwhatever docs users chuck at itβand spit out real answers. No coffee breaks, no βfile not found,β just 24/7 research slave energy. Multi-file? Easy. Context that actually makes sense? Also yes.
- Drag-and-drop any doc β instant summary
- Photographic memory on steroids
- Sleep? Never heard of her
#AI #Grok #TechLife #NoMoreManualReading
π @missedminutes π
β‘1
βοΈ Microsoft drops call-center-ai: 1 API call = your new AI intern on the phone ππ€
β¨ Key Points:
π Links:
github.com/microsoft/call-center-ai
#AI #Tech #MachineLearning #Future
π @missedminutes π
β¨ Key Points:
- Drop an API β AI agent dials out, small-talks, takes notes, hangs up. No coffee breaks.
- Give it a real phone number; customers can ring the bot directly. 24/7 shift coverage without HR crying.
- Open-source, Python-ready, Azure-flavoured. Fork, roast, deploy.
- Want to kill hold music forever? PRs welcome.
π Links:
github.com/microsoft/call-center-ai
#AI #Tech #MachineLearning #Future
π @missedminutes π
GitHub
GitHub - microsoft/call-center-ai: Send a phone call from AI agent, in an API call. Or, directly call the bot from the configuredβ¦
Send a phone call from AI agent, in an API call. Or, directly call the bot from the configured phone number! - microsoft/call-center-ai
β‘1
βοΈ Grok Just Ate Every Benchmark for Breakfast π³
While devs were doom-scrolling, Grok quietly speed-ran the leaderboard. November stats? Total sweep: BlackBox, Terminal-Bench, GPQA Diamond, SciCode, Roo, Kilo, Clineβ¦ name it, itβs #1.
On OpenRouter itβs not just βpopularββitβs the token-guzzling champ of today, this week, this month. Python, JS, Java, C++, SQL? All its playground.
#AI #Grok #xAI #CodeLife
π @missedminutes π
While devs were doom-scrolling, Grok quietly speed-ran the leaderboard. November stats? Total sweep: BlackBox, Terminal-Bench, GPQA Diamond, SciCode, Roo, Kilo, Clineβ¦ name it, itβs #1.
On OpenRouter itβs not just βpopularββitβs the token-guzzling champ of today, this week, this month. Python, JS, Java, C++, SQL? All its playground.
Translation: if your IDE still auto-completes like itβs 2022, congrats on the vintage aesthetic. Everyone else: maybe time topip install grok?
#AI #Grok #xAI #CodeLife
π @missedminutes π
β‘1
jamesdouma battleangelviv You can feel the sentience of the car growing. It is palpable.
π @missedminutes π
π @missedminutes π
β‘1
Forwarded from Data eXplore β Data Science, ML, Big Data, LLMs
Basic database concepts you should learn if you really want to understand scaling:
SAVE this
π€ Data Science, ML & Big Data with @DataXplore
SAVE this
B+ trees
LSM trees
Write-Ahead Logging (logging before writing)
Two-phase commit
Three-phase commit
Read-only replicas
Leader-follower replication
Sharding / partitioning
Query caching
Secondary indexes
Vector indexes (Vector Indexes, e.g., FAISS, HNSW)
Distributed JOINs
Materialized views
Event Sourcing (event-based state storage)
Change Data Capture (tracking data changes)
π€ Data Science, ML & Big Data with @DataXplore
βοΈ Grok 4 Fast: The AI Wallet Saver
Devs still sleeping on Grok 4 Fast? π
Same IQ as GPT-5, price of a chai. 95 % off = 100 % stonks.
#AI #Tech #MachineLearning #Future
π @missedminutes π
Devs still sleeping on Grok 4 Fast? π
Same IQ as GPT-5, price of a chai. 95 % off = 100 % stonks.
π€ Broke vs. Woke:
GPT-5 β $10 / M out
Grok 4 Fast β $0.5 / M out
Math ainβt mathing? exactly.
β‘οΈ Speed run: 344 t/s vs GPT-5βs 138. Compile before your coffeeβs hot.
TL;DR: 98 % cheaper, 2.5Γ faster, zero cap.
Still burning cash on big-name APIs? yikes.
#AI #Tech #MachineLearning #Future
π @missedminutes π
β‘2
β‘2
Data eXplore β Data Science, ML, Big Data, LLMs
MIRA by #ByteDance
Multimodal Imagination for Reasoning Assessment, a test that measures Reasoning models need not only words but also images
π What is the idea?
- Where text doesn't help, images sharply improve the model's thinking.
- If the model is given drawings of intermediate steps, accuracy on average increases by 33.7%.
- The benchmark includes 546 tasks in 20 categories where you need to "see" rather than just read: blocks, mirrors, trajectories, forces, etc.
How the test is structured?
- direct question
- reasoning with text
- reasoning with visual steps (sketches)
Multimodal Imagination for Reasoning Assessment, a test that measures Reasoning models need not only words but also images
π What is the idea?
- Where text doesn't help, images sharply improve the model's thinking.
- If the model is given drawings of intermediate steps, accuracy on average increases by 33.7%.
- The benchmark includes 546 tasks in 20 categories where you need to "see" rather than just read: blocks, mirrors, trajectories, forces, etc.
How the test is structured?
- direct question
- reasoning with text
- reasoning with visual steps (sketches)
Text alone often worsens performance because words poorly describe space.
If the model is given images, the result improves significantly, especially in the exact sciences.
π @missedminutes π
If the model is given images, the result improves significantly, especially in the exact sciences.
π @missedminutes π
β‘1
Forwarded from Data eXplore β Data Science, ML, Big Data, LLMs
This media is not supported in your browser
VIEW IN TELEGRAM
An open project aimed at making 100 million scientific articles accessible through LLM-generated structured summaries.
π§ A dataset of 100,000 summaries
𧩠Two fine-tuned LLM models for analysis and structuring
π A 3D visualizer that shows the relationships between scientific papers
Models & Visualizer
π€ Data Science, ML & Big Data with @DataXplore
π§ A dataset of 100,000 summaries
𧩠Two fine-tuned LLM models for analysis and structuring
π A 3D visualizer that shows the relationships between scientific papers
Models & Visualizer
π€ Data Science, ML & Big Data with @DataXplore
SanDisk announced π΄ the smallest SSD , called SanDisk Extreme Fit :
β Weighs only 3gm
β It is so small that you can keep it connected with any device & work , that's why it is called "Plug & Play"
β It uses USB 3.2 Gen 1 port
β Maximum speed : 400MB/S
β Upto 1TB storage
π @missedminutes π
β Weighs only 3gm
β It is so small that you can keep it connected with any device & work , that's why it is called "Plug & Play"
β It uses USB 3.2 Gen 1 port
β Maximum speed : 400MB/S
β Upto 1TB storage
π @missedminutes π
β‘1
This new UI is FIRE @GeminiApp ! β¨
Pls roll out this on all accounts lmao, in my principal acc, I dont have it π
In moment, Google cooking really hard, end of this year will be beautiful :D
π @missedminutes π
Pls roll out this on all accounts lmao, in my principal acc, I dont have it π
In moment, Google cooking really hard, end of this year will be beautiful :D
π @missedminutes π
β‘1