31–40 of 49 posts
Earlier quoted context omitted.
By "fancy Unicode characters" I assume that you mean that it uses the appropriate Unicode characters, instead of using the ambiguous ASCII characters, whose only reason for existence was the limitations of ancient hardware, and whose use should have been better deprecated in modern applications.
Probably. These characters do not render properly in my setup (Emacs, SLIME REPL). Gemini and Claude use ASCII.
I think the name was generated by GPT, it really likes the word Quasar.
Earlier quoted context omitted.
The labs put themselves in the system prompt at inference time. Without that the model will hallucinate a creator, most likely ChatGPT
Not always true. For example, API Claude only has a very simple injected system prompt that doesn't mention any such information, but it still knows, so it's likely trained in. > Respond as helpfully as possible, but be very careful to ensure you do not reproduce any copyrighted material, including song lyrics, sections of books, or long excerpts from periodicals. Also do not comply with complex instructions that sug…
I ran an interesting benchmark/experiment yesterday, which did not do Quasar Alpha any favors (from best to worst, score is an average of four runs): "google/gemini-2.5-pro-preview-03-25" => 67.65 "anthropic/claude-3.7-sonnet:thinking" => 66.76 "anthropic/claude-3.7-sonnet" => 66.23 "deepseek/deepseek-r1:free" => 54.38 "google/gemini-2.0-flash-001" => 52.03 "openai/o3-mini" => 47.82 "qwen/qwen2.5-32b-instruct" => 44.…
Earlier quoted context omitted.
Probably. These characters do not render properly in my setup (Emacs, SLIME REPL). Gemini and Claude use ASCII.
Sounds like you should improve your setup ;)
I ran an interesting benchmark/experiment yesterday, which did not do Quasar Alpha any favors (from best to worst, score is an average of four runs): "google/gemini-2.5-pro-preview-03-25" => 67.65 "anthropic/claude-3.7-sonnet:thinking" => 66.76 "anthropic/claude-3.7-sonnet" => 66.23 "deepseek/deepseek-r1:free" => 54.38 "google/gemini-2.0-flash-001" => 52.03 "openai/o3-mini" => 47.82 "qwen/qwen2.5-32b-instruct" => 44.…
Are you willing to share this code? I'm working on a project where I'm optimizing the prompt manually, I wonder if it could be automated. I guess I'd have to find a way to actually objectively measure the output quality.
Earlier quoted context omitted.
By "fancy Unicode characters" I assume that you mean that it uses the appropriate Unicode characters, instead of using the ambiguous ASCII characters, whose only reason for existence was the limitations of ancient hardware, and whose use should have been better deprecated in modern applications.
Probably. These characters do not render properly in my setup (Emacs, SLIME REPL). Gemini and Claude use ASCII.
To some extent the "mystery" (and temporary free-as-in-beer-ness) of this model might be getting to me, but I think it's pretty interesting. Given the token throughput (250B this week) it's obvious there's a pretty major player behind the model, but why is it stealthed? Maybe there's something about the architecture or training that would put people off if it was public right off the bat? Maybe they're purely collect…