To be clear, this isn’t a test of the LLMs themselves, but their presentation from the vendor’s most consumer-ish UI. I assume many of these are getting the answer in the system prompt. My point is just it isn’t done consistently and I liked DeepSeek bothered.
LLMS
-
SemiKong: First Open Source Semiconductor-Focused LLM Model
By
–
SemiKong, a model built with Llama, is the world's first open source semiconductor-focused LLM. With this work @aitomatic is enabling semiconductor companies to build Domain-Expert Agents to capture and scale their deep domain expertise https://
go.fb.me/9yq4mq -
GPT-4o Model Parameters and Size Specifications
By
–
do you happen to know the number params/ size of 4o?
-
Data Quality Importance Underestimated in LLM Training
By
–
I'm not bearish on them.. just saying that people underestimate the opportunity cost of a good data mixture for LLMs
-
Parameter Count Estimates: Claude Sonnet 3.5 vs OpenAI o1 o3
By
–
what's the current best estimate of number of parameters for claude sonnet 3.5 and OAI o1/ o3? leaks/ guesstimates all welcome
-
Data Quality Limitations in AI Model Development
By
–
if they could've, they would've, provided they have enough high quality data as DS – twitter slop will only go as far
-
DeepSeek-V3 Technical Paper Released
By
–
read the paper https://
github.com/deepseek-ai/De
epSeek-V3/blob/main/DeepSeek_V3.pdf
… -
DeepSeek V3: China’s AI Model Rivals GPT-4o with Less Compute
By
–
It is quite fitting that DeepSeek, China’s leading LLM lab, releases its latest model V3 on Christmas.
– on-par with GPT-4o & Claude 3.5 Sonnet
– trained w/10x less compute The bitter lesson of Chinese tech: they work while America rests, and catch up cheaper, faster & stronger -
ChatGPT struggles to identify its own model
By
–
ChatGPT is surprisingly bad at knowing what model it’s using — it either confidently gives an incorrect version or at best it’s vague (“based on the GPT-4 architecture”). No current model seems to ever know its exact name.
-
Gemini model naming behavior when specifying “model”
By
–
Gemini thinks you’re talking about the Gemini Advanced frontend unless you add “model,” which is fair. Once you do, all except for one experimental model (1206) can correctly state their name as shown in the UI: