New GPT-4 Turbo now available in GA with built-in vision capabilities! You can now use function calling and JSON mode with image inputs—something many of you have been asking for. Happy building!
LLMS
-
Building Interactive Prototypes with LLM Assistance: Full Process
By
–
This is a great video for watching someone who builds quick interactive prototypes using LLM assistance walk through their full process
-
Large Language Model Token Processing Performance Reliability
By
–
Interesting! I think that might be a bug on your end (I’ve heard it can be really buggy for some people in preview) I’ve done over 700,000 tokens and it worked like a charm
-
RecurrentGemma Open Source: Griffin Achieves Inference Throughput Gains
By
–
I’m very proud of our team for open sourcing RecurrentGemma. Yes, recurrence is back and it results in huge gains at inference time. Just look at the impressive throughput plot below. For details, please see the paper and GitHub page: Griffin: Mixing Gated Linear Recurrences
-
Microsoft Provides Free GPT-4 Level Model for Global Access
By
–
Also kudos to Microsoft for providing the only free GPT-4 level model available worldwide. Many people could not afford access to a frontier model otherwise. I hope they keep up that commitment.
-
Gemini 1.5 Pro: Accessing 1M+ Context Token Length
By
–
Step 4:
— Rowan Cheung (@rowancheung) 9 avril 2024
In Google AI Studio, just go to the Model settings on the right, and make sure you've selected Gemini 1.5 Pro over Gemini 1.0 Pro.
That's it! Start prompting away with your 1M+ context token context length. pic.twitter.com/Z19RClkpqbStep 4: In Google AI Studio, just go to the Model settings on the right, and make sure you've selected Gemini 1.5 Pro over Gemini 1.0 Pro. That's it! Start prompting away with your 1M+ context token context length.
-
Gemini 1.5 Capabilities: Video Analysis, Translation, Understanding
By
–
First, in case you missed it, here's my thread on Gemini 1.5 and its use cases. Some capabilities I covered: -Breaking down + understanding a movie
-Translating text into a rare language
-Watching an AI video and determining if it's AI or not -
Google Gemini 1.5 Pro’s Million Token Context Length
By
–
Google Gemini 1.5 Pro's 1,000,000+ token context length is not talked about enough.
— Rowan Cheung (@rowancheung) 9 avril 2024
At launch, 1.5 Pro was overshadowed by OpenAI's Sora and only accessible to select users.
But now, anyone can try it for free.
Here's a quick tutorial 🧵: pic.twitter.com/lIjeHE8W9uGoogle Gemini 1.5 Pro's 1,000,000+ token context length is not talked about enough. At launch, 1.5 Pro was overshadowed by OpenAI's Sora and only accessible to select users. But now, anyone can try it for free. Here's a quick tutorial :
-
LangSmith Evaluations: Choosing Right Prompts and LLM Models
By
–
LangSmith Evaluations With the rapid pace of AI, developers are often faced with a paradox of choice: how to choose the right prompt, how to trade-off LLM quality vs cost? Evaluations can accelerate development with structured process for making these decisions. But, we've
-
Assessing Language Model Persuasiveness: Ethical Challenges
By
–
Assessing the persuasiveness of LMs is inherently difficult. Persuasion is a nuanced phenomenon shaped by many subjective factors, and is further complicated by the bounds of ethical experimental design. We detail the challenges we encountered so others can build on our work.