simonbutt

6 ms·

1.
by simonbutt 3y ago
A couple weeks ago I rebuilt the https://hiresynth.ai landing page and docs using vitepress and was super impressed with the speed of implementation and how pretty it was by default. Would recommend for any text heavy static sit
2.
by simonbutt 3y ago
While opensource models allow you to explore wider topics, I wonder how well mixtral holds context on longer exchanges?
3.
by simonbutt 3y ago
Wrote a post on multimodal LLMs (GPT-4V, Gemini Ultra, Claude3 Opus) attempting the puzzles at https://hiresynth.ai/blog/googleio_puzzle_multimodal_eval.ht... . Seems very easy for humans, far harder for LLMs currently
4.
by simonbutt 3y ago
GPT-4V, Claude 3 Opus and Gemini Ultra go head to head in solving GoogleIO Puzzle 2024
5.
Evaluating Multimodal LLMs Using the Google IO 2024 Puzzle(hiresynth.ai)
3 points by simonbutt 3y ago|2 comments
6.
by simonbutt 3y ago
This feels like the start of a lawsuit. If Tesla is state of the art and requires 5x less funding per charger than competition; what is the reasoning behind picking the competitors?
7.
Am I using it wrong or has the ChatGPT Bing plugin degraded?(chat.openai.com)
1 points by simonbutt 3y ago|1 comments
8.
by simonbutt 3y ago
This feels as if they've tacked the bing "chat with gpt4" to chatgpt while the previous version was able to browse multiple places and provide (usually) decent responses. Have you found any custom instructions to improve the
9.
by simonbutt 3y ago
Once in a while HN provides absolute gems. Would never have known about this anti procrastination feature otherwise!
10.
by simonbutt 3y ago
Logic for azure chatgpt's "infinite context" summarisation is in https://github.com/microsoft/azurechatgpt/blob/main/src/feat... *Edit Azure chatgpt, would be amazed/disappointed
11.
by simonbutt 3y ago
Google released PaLM pricing per character at $0.0005/1k (for chat, instruct is 0.001). With OpenAI tokens at approx 4 chars per token, this matches $0.002/1k tokens for 3.5. Let's hope OpenAI retaliate, cut prices to maint
12.
Google PaLM Pricing matches gpt3.5 for chat(cloud.google.com)
1 points by simonbutt 3y ago|4 comments
13.
by simonbutt 3y ago
That first photo has to be Midjourney. Have we hit the turning point where AI art is suddenly embedded in everything?
14.
by simonbutt 3y ago
You won't catch everything but I've found if you specify that it's okay to respond that it doesn't know the answer and essentially accuracy on a smaller subset of answers > responding to all, it often will say it does
15.
by simonbutt 3y ago
Q: "Tell me about Stonks GPT?" A: "Stonks GPT is a trading name for Stocks & Breads, a UK-based provider of bread and breakfast products. The company was founded in 1894 and is headquartered in London. It is a leading pro
16.
by simonbutt 3y ago
Product market fit