I've been switching up to Sonnet when Haiku doesn't quite give me good enough results, it's a nice mix of speed and quality
LLMS
-
Balancing Model Performance and Latency for User-Facing Features
By
–
I'm considering it for user-facing features where latency performance is important but Haiku isn't quite giving me good enough results
-
Choosing Between Opus and GPT-4 for Production Features
By
–
For my own personal use, I tend to stick with the best available model, so Opus or GPT-4 But now that I'm building user-facing features on them both cost and speed are more important, so I'm being more thoughtful about which model to use
-
OpenAI’s missing middle tier between GPT-3.5 and GPT-4
By
–
Surprisingly the OpenAI lineup seems to be missing that middle piece – something that sits between the cheap and fast gpt-3.5-turbo and their GPT-4 class models The difference between GPT-4 and gpt-4-turbo doesn't quite feel like Opus to Sonnet
-
GPT-4 Already Making Serious Impact in Law, Medicine, Business
By
–
There is a ton of debate about how good AI might get, but not enough recognition that the research on AI in law, medicine, business etc. finds that GPT-4 class AI is good enough to make a serious difference in work & education In some ways, future capabilities are a distraction
-
Three Sizes of AI Models: Haiku, Sonnet, and Opus Comparison
By
–
Something I have learned from Claude 3 is that I really like my models in three sizes: the fast one (Haiku), the slow but "best" one (Opus) and the hard to define but spectacularly useful one in the middle (Sonnet) Mistral 7B / Mixtral 8x7B / Mixtral 8x22B feels similar to that
-
GPT-4 128k Context Window Now Available for Prompt Bots
By
–
New: GPT-4 128k is now available as a base for prompt bots! The longer context window coupled with the power of the latest GPT-4 update should enable a variety of new use cases. Enjoy! pic.twitter.com/rKsHPQj1mD
— Poe (@poe_platform) 16 avril 2024New: GPT-4 128k is now available as a base for prompt bots! The longer context window coupled with the power of the latest GPT-4 update should enable a variety of new use cases. Enjoy!
-
Claude 3 Opus Now Available on Amazon Bedrock
By
–
Our most capable model, Claude 3 Opus, is now generally available on Amazon Bedrock.
— Anthropic (@AnthropicAI) 16 avril 2024
Alongside Sonnet and Haiku, Opus provides businesses with exceptional intelligence, fluency, and reasoning capabilities.
Get started today: https://t.co/Pa4YEyUK4i https://t.co/1bteaa6zWZOur most capable model, Claude 3 Opus, is now generally available on Amazon Bedrock. Alongside Sonnet and Haiku, Opus provides businesses with exceptional intelligence, fluency, and reasoning capabilities. Get started today: https://
aws.amazon.com/bedrock/claude/ -
Startups Share Technical Stack Details for LLM Implementation
By
–
Startups usually don’t say a lot about how their stack works (or it’s just a wrapper around another company’s LLM) so it’s great to get this detailed deep dive
-
Watson’s Jeopardy Victory Parallels Modern AI Language Models
By
–
5 paras from @JohnMarkoff01
's Feb 16, 2011, NYT story on Watson when it won Jeopardy. Substitute ChatGPT, or similar, for Watson and one of many companies for IBM and it still reads pretty well.