Tencent's EVIE Model Tops ViDoRe Benchmarks, Cuts Vector Storage Costs by 32x
jacek2023 · reddit · 2026-08-17
Tencent released EVIE-Preview-4.5B, a multilingual Visual Document Retrieval (VDR) model built upon Qwen3.5-4B. It employs a ColBERT-style late interaction mechanism with native 128-dimensional multi-vector token embeddings, achieving SOTA results on ViDoRe V1+V2 and V3 benchmarks while outperforming larger 8B models. The ultra-compact 128D index design reduces vector storage and indexing costs by 8x to 32x, supporting zero-shot generalization across multiple languages and visual formats like charts and tables.
Related event: Tencent's EVIE Tops ViDoRe Benchmark with 32x Storage Savings(3 posts)→
More from Models
- GLM 5.5 rumored to drop soon, coding ability expected to match frontier models — bindureddy · 2026-10-16
- Five rounds of adversarial prompting make Opus 5.5 and Argon converge on identical output — LanceCharger · 2026-10-16
- DIY benchmark ranks 18 models on subjective judgments, GLM-5.3 tops consensus match — arkuto · 2026-10-16
- Bindu Reddy: Google's Argon excels at complex doc writing at half the cost of Opus — bindureddy · 2026-10-16
- MAI-2 is 2x bigger than V4-Pro while V4.1F is 3x smaller: Microsoft's scaling ladder questioned — teortaxesTex · 2026-10-16
- Step 5 Preview hits DeepInfra day-zero: 600B MoE, 1M context, $1/M input tokens — StepFun_ai · 2026-10-16