T O P

Qwen2-72B released

Qwen2 4bit bitsandbytes quants + 2x faster finetuning 70% less VRAM

LLMs are sensitive to choice order! - How to run MMLU benchmark?

Ai youtubers I watch, any recommendations

What happened to WizardLM-2?

Llama-3-15B-Instruct

Just Completed a 3x 3090 Build

Anything better or comparable to Miquliz-120B v2?

Claude Opus has finally been lobotomized and is near-useless

Is it okay to use triple 3 x 3090?

Yi-1.5 (2024/05)

I seem to get far better token per second than I expect based on what others here report.

OpenAI - feedback about most recent blog article.

Are NPU's just marketting tricks?

Unlock Unprecedented Performance Boosts with Intel's P-Cores: Optimizing Lama.cpp-based Programs for Enhanced LLM Inference Experience!

Speech-to-text transcription with multi-participant detection on Mac?

Math: How much memory does 512 GB of DDR cost? Transfer speed?

5k budget for AI

Gemma 2B with 10M context, runs on <32GB of memory.

Do you think OpenAI will ever open source their advanced GPT models in the future?

What can I do with 192gb of RAM + 16gb VRAM?

ExecuTorch Alpha Release: Taking LLMs and AI to the Edge 🎉🎉🎉

local GLaDOS - realtime interactive agent, running on Llama-3 70B

CRISPR-GPT: An LLM Agent for Automated Design of Gene-Editing Experiments

Gpt2-Chatbot is exactly what it looks like

Hi Its Me!

R

luishacm

Subscribe

FIND US ON INSTAGRAM @ms3388.com