Also as some ppl pointed out, this is harder to get right so quality varies across solutions and none working perfectly yet.* People don’t want to just find specific pieces of data, but want summaries and patterns. This is more than basic RAG. *tell me if you find one!
LLMS
-
Comparison of Generative AI Model Capabilities
By
–
There was no way for gen2 iirc. I don't expect it to be there for gen3
-
Haiku 3.5 Expected to Be Best Small Model
By
–
I expect Haiku 3.5 to be really really good. Maybe the best small model
-
Claude Excels at Following Instructions Despite Style Limitations
By
–
Interestingly enough, Claude also loses points on writing style and formatting vs other top models I think this is somewhat contrary to public opinion, but Claude is just much better at following instructions
-
GPT-4o vs GPT-4 Turbo: Choosing the Right Model for Code
By
–
Depending on your code language btw, you might by want to still use GPT 4o or GPT 4 Turbo
-
Claude 3.5 Sonnet Tops SEAL Evals in Instruction Following Coding
By
–
We re-ran SEAL evals on the new @AnthropicAI Claude 3.5 Sonnet model. It is now:
– #1 on Instruction Following
– #1 on Coding Congratulations to Anthropic on a great new model! P.S. we’re adding new evals to SEAL, so if you have an idea for an eval, let us know below -
How to test the TestingCatalog GPT
By
–
Here is how to test it:
– Start a new chat with TestingCatalog GPT
– Use a starter prompt or ask “whats new?”
– Select any number to expand 1-5 – Ask “show it to me” Just tested on the Android beta app and it works there as well -
GEN-3 Alpha Now Available for Everyone
By
–
🔴 ¡GEN-3 ALPHA DISPONIBLE PARA TODOS! https://t.co/uJRif5DK6J
— Carlos Santana (@DotCSV) 1 juillet 2024¡GEN-3 ALPHA DISPONIBLE PARA TODOS!
-
How to use the TestingCatalog GPT for latest updates
By
–
Here is how to use it: – start a new chat with TestingCatalog GPT
– use a starter prompt or ask “whats new?”
– select any number to expand 1-5 – ask “show it to me” -
Anthropic Deciphers Claude’s Internal Patterns Breakthrough
By
–
¡¡NUEVO VÍDEO!! ¿Creéis eso de que las IAs son cajas negras que no podemos entender? ¡Pues la cosa ha cambiado! Y un trabajo reciente de Anthropic ha permitido descifrar muchos de los patrones internos de su modelo Claude! Ah… y volver LOCA a la IA