SPLADE's original author releases new sparse embedding model: 150M params beat models up to 1B
IgorCarron · x · 2026-09-18
The original author of the SPLADE line (widely regarded as the definitive sparse embedding architecture) has released the biggest advancement in sparse embedding models since SPLADE-v3: the new model is only 150M parameters yet beats every sparse model up to 1B at search, and reaches sub-millisecond latency using Seismic indexing.
More from Models
- Stanford's 10-person Marin open lab is live-training a 535B model in the open — wandb · 2026-09-18
- Self-Proclaimed ChatGPT Co-Inventor Launches Jev, Claims 200x Speed at 1/400 Cost — iamrobotbear · 2026-09-18
- Codex Pro User Says Usage Limits Got 5-10x Worse, Can't Even Buy Another Plan — Junra · 2026-09-18
- GPT-6 Astra Deciphers an Undeciphered 1918 German WWI Radio Transmission — moultano · 2026-09-18
- Sakana AI Introduces Fugu Max and Fugu Ultra v2 Models — SakanaAILabs · 2026-09-18
- Gemini 3.8 Live Architecture Breakdown: Sub-100ms Native Audio and Real-Time Tool Calling — 4bTechDecode · 2026-09-18