Blaxel is joining Baseten. Read more

Pankaj Gupta

Co-Founder

Model performance

How we built production-ready speculative decoding with TensorRT-LLM

Pankaj Gupta
2 others
Model performance

A quick introduction to speculative decoding

Pankaj Gupta
2 others
Infrastructure

Evaluating NVIDIA H200 Tensor Core GPUs for LLM inference

Pankaj Gupta
1 other
Model performance

How to serve 10,000 fine-tuned LLMs from a single GPU

Pankaj Gupta
1 other
Infrastructure

Using fractional H100 GPUs for efficient model serving

Matt Howard
3 others
Model performance

Benchmarking fast Mistral 7B inference

Abu Qader
3 others
Model performance

33% faster LLM inference with FP8 quantization

Pankaj Gupta
1 other
Model performance

FP8: Efficient model inference with 8-bit floating point numbers

Pankaj Gupta
1 other
Model performance

40% faster Stable Diffusion XL inference with NVIDIA TensorRT

Pankaj Gupta
2 others
123

Explore Baseten today

Start deployingTalk to an engineer