Guillaume Lample
Co-Founder & Chief Scientist
Verified Proof of Work Artifacts
2 items catalogedEach artifact below represents an authenticated research publication, production code repository, or technical architectural framework directly authored or co-created by Guillaume Lample. Every entry undergoes editorial source verification.
Mixtral of Experts: Sparse Mixture of Experts with Open Weights
Co-authored the paper introducing Mixtral 8x7B, a decoder-only model where each layer is composed of 8 feedforward blocks with a router network that selects 2 experts per token, outperforming Llama 2 70B while maintaining the inference speed of a 12B model.
MoE models require substantial VRAM for weight storage despite sparse compute activations during generation.
LLaMA: Open and Efficient Foundation Language Models
Core co-author of the original Meta LLaMA 7B-65B release, proving that smaller models trained on significantly more tokens than Chinchilla recommendations surpass larger architectures and enable democratized research.
Base pre-trained checkpoints require safety and instruction fine-tuning to prevent conversational misalignment.