generally, firmware is a lot more difficult to open-source because there might be IP in there that the company might not own (of vendor chips that are on the GPU board, etc.)
@soumithchintala
-
PyTorch Pay-Per-Use Model Through Advanced Compiler Technology
By
–
I think some day, projects like PyTorch will make users pay only for the overhead they use. Lots of R&D needed to get there: compiler technology, type systems, various other things. Actually, LuaJIT was really cool in that aspect — one of the most brilliant pieces of
-
Growing Availability of Open-Source AI Models Benefits Industry
By
–
+1 the more open-models available at this level, the better
-

Llama3-70B Ranks 5th in LMSys Arena Early Results
By
–
very early LMSys Arena results peg llama3-70B at 5th place (the variance is still pretty high, so it can jump up or down a bit).
This is so exciting.
Can't wait to see how the 405B fares once it is released. https://
chat.lmsys.org/?leaderboard -
Popular LLM Training Frameworks: Hugging Face Transformers Axolotl UnSloth
By
–
in the community: huggingface transformers, axolotl, unsloth, lit-gpt are popular options.
-
Meta Releases Live Preview and Animation Features for Image Generation
By
–
There's another quieter release from @AIatMeta today that's really cool.
* Live Preview: As you type your image prompt, you get a live preview, making iterating for a good image easier.
* Animate: now you can animate images for short bursts -
Memory and Battery Limitations in Laptop Development
By
–
i find memory and battery to be a limiting factor for laptop development. i'm still stuck on an M1 macbook with 16GB memory, and once i start doing computing on the laptop battery depletes too quickly (like on a flight and stuff).
of course i can buy one of those ultra macbooks, -
Gemini’s Published Content Gap: A Tech Industry Observation
By
–
as the semi-serious joke goes: if they published it, its not in gemini
-
Generation Workload TCO Trade-off Analysis
By
–
for the workloads we were targeting with this generation, was a better TCO trade-off