Axel 👨💻 Developer
@axelgarciak.bsky.social
👨💻 Software Engineer 💾 Software minimalist/retro
🤖 AI tinkerer 🏗️ Building tech communities
🇪🇺 UK 🇬🇧🇩🇪🇻🇪 | Check bio: axelgarciak.com/bio
🤖 AI tinkerer 🏗️ Building tech communities
🇪🇺 UK 🇬🇧🇩🇪🇻🇪 | Check bio: axelgarciak.com/bio
August 16, 2025 at 7:31 PM
Many good releases from open and closed-source providers. I'm eagerly waiting for what DeepSeek is going to release.
May 1, 2025 at 4:45 PM
Many good releases from open and closed-source providers. I'm eagerly waiting for what DeepSeek is going to release.
They are good! Obviously not what you get from models like O4-mini, Claude 3.7 or Gemini 2.5 Pro.
But they are great for local use cases.
It's nice that they are hybrid, i.e: thinking can be deactivated.
Qwen2.5 was already the best in its class for a while! Hopefully we'll get a Coder fine-tune!
But they are great for local use cases.
It's nice that they are hybrid, i.e: thinking can be deactivated.
Qwen2.5 was already the best in its class for a while! Hopefully we'll get a Coder fine-tune!
April 30, 2025 at 11:22 PM
They are good! Obviously not what you get from models like O4-mini, Claude 3.7 or Gemini 2.5 Pro.
But they are great for local use cases.
It's nice that they are hybrid, i.e: thinking can be deactivated.
Qwen2.5 was already the best in its class for a while! Hopefully we'll get a Coder fine-tune!
But they are great for local use cases.
It's nice that they are hybrid, i.e: thinking can be deactivated.
Qwen2.5 was already the best in its class for a while! Hopefully we'll get a Coder fine-tune!
Yeah benchmarks don't translate fully to real use cases. Claude is still king for coding.
March 12, 2025 at 8:21 PM
Yeah benchmarks don't translate fully to real use cases. Claude is still king for coding.
Of course benchmarks will differ from real life use cases.
However, previous Gemma models were quite good relative to their size, so I'm sure Gemma 3 is really good!
However, previous Gemma models were quite good relative to their size, so I'm sure Gemma 3 is really good!
March 12, 2025 at 7:49 PM
Of course benchmarks will differ from real life use cases.
However, previous Gemma models were quite good relative to their size, so I'm sure Gemma 3 is really good!
However, previous Gemma models were quite good relative to their size, so I'm sure Gemma 3 is really good!
It has been happening for a while before ChatGPT came out, but you're right in that it will accelerate.
January 19, 2025 at 8:59 PM
It has been happening for a while before ChatGPT came out, but you're right in that it will accelerate.
It is useful indeed, I use it every day, but it's far from being AGI or ASI.
January 19, 2025 at 8:58 PM
It is useful indeed, I use it every day, but it's far from being AGI or ASI.
Debian or Fedora if you want mainstream distros.
January 14, 2025 at 4:48 AM
Debian or Fedora if you want mainstream distros.
I like that AMD has technologies that are more widely applicable, such as FSR, but as usual NVIDIA's is on the lead with the DLSS 4 technology.
January 13, 2025 at 9:45 PM
I like that AMD has technologies that are more widely applicable, such as FSR, but as usual NVIDIA's is on the lead with the DLSS 4 technology.
Yeah frame interpolation is also cool. The key difference is that interpolation is reactive (analyzing past frames), DLSS is predictive (anticipating future frames).
January 13, 2025 at 9:41 PM
Yeah frame interpolation is also cool. The key difference is that interpolation is reactive (analyzing past frames), DLSS is predictive (anticipating future frames).
Gotta account for those days when the OpenAI API is down 😏
January 12, 2025 at 3:18 PM
Gotta account for those days when the OpenAI API is down 😏
Dave is a 10xer.
January 12, 2025 at 9:46 AM
Dave is a 10xer.
Agreed. To their credit, these newer GPUs have fp4 support whereas before you had to run at fp8 even when running fp4 calculations.
If you were to use fp4 exclusively, there is an advantage in speed from these newer GPUs.
If you use fp8 there is still an improvement but far less than advertised.
If you were to use fp4 exclusively, there is an advantage in speed from these newer GPUs.
If you use fp8 there is still an improvement but far less than advertised.
January 7, 2025 at 2:05 PM
Agreed. To their credit, these newer GPUs have fp4 support whereas before you had to run at fp8 even when running fp4 calculations.
If you were to use fp4 exclusively, there is an advantage in speed from these newer GPUs.
If you use fp8 there is still an improvement but far less than advertised.
If you were to use fp4 exclusively, there is an advantage in speed from these newer GPUs.
If you use fp8 there is still an improvement but far less than advertised.