Sublime

Wake up babe new SOTA quantization method just dropped!!!! TEQ (Trainable Equivalent Transformation) preserves FP32 precision of the model output while using low-precision quantization. Training requires only 1K steps and < 1% of the original model’s trainable parameters. Results are on par... See more

Understanding CPU vs Memory Patterns in Modern but Frugal 💰 Data Engineering: An Infrastructure Perspective - DuckDB: Efficient memory use with spikes during complex joins. Excellent for data quality checks, handling 70GB Parquet analytics with 32GB RAM. - Polars: 10x better... See more

Linear scaling achieved with multiple DeepSeek v3.1 instances. 4x macs = 4x throughput. 2x M3 Ultra Mac Studios = 1x DeepSeek @ 14 tok/sec 4x M3 Ultra Mac Studios = 2x DeepSeek @ 28 tok/sec DeepSeek V3.1 is a 671B parameter model - so at its native 8-bit... See more

eXmY: A Data Type and Technique for Arbitrary Bit Precision Quantization https://t.co/NyV4BNPfBO

Using light as a neural network, as this viral video depicts, is actually closer than you think. In 5-10yrs, we could have matrix multiplications in constant time O(1) with 95% less energy. This is the next era of Moore's Law. Let's talk about Silicon Photonics... 1/9 https://t.co/vn9K47xDvU