Releases·Model Release
Cohere launches Embed 5 Pro and Fast, embedding models with shared embedding space and RCP-nDCG@10 eval
Cohere launched Embed 5, a two-model embeddings family: Embed 5 Pro, which posts the top average score on ViDoRe V3 ahead of Voyage 4, Gemini Embedding 2, and Jina Embeddings v5, and Embed 5 Fast, a lighter low-latency model for high-volume use. Both models share a single embedding space, so teams can index with Pro and query with either without re-indexing. Embed 5 is also Cohere's first family evaluated with its new RCP-nDCG@10 retrieval methodology.
2 engagement1 sourceWed, Sep 30, 2026, 01:20 PM
AI Overview
- Cohere launches Embed 5 in two models: Embed 5 Pro, top average score on ViDoRe V3 ahead of Voyage 4, Gemini Embedding 2, and Jina Embeddings v5
- Embed 5 Pro is trained on 100+ languages and targets enterprise documents like SEC filings and slide decks
- Embed 5 Fast outperforms Voyage 4 nano and Qwen3-VL-Embedding-2B at lower latency and cost
- Both Embed 5 models share a single embedding space, letting users index with Pro and query with either without re-indexing
- Embed 5 is the first model family evaluated with RCP-nDCG@10, Cohere's latest retrieval quality methodology
- Cohere will detail Parse 5, Embed 5, and its search roadmap in a webinar on October 8 on X Live
Related
Releases·Model Release
Tencent releases EVIE-8B and EVIE-4.5B visual document retrieval models, topping ViDoRe V3 at 66.75 nDCG@10
Tencent released EVIE-8B and EVIE-4.5B on Hugging Face, models for high-capacity visual document retrieval. EVIE-8B reports state-of-the-art retrieval performance with 66.75 nDCG@10 on ViDoRe V3 and 4096D per-token multi-vector embeddings that preserve layout, typography, charts, and tables.
Tencent released EVIE-8B and EVIE-4.5B on Hugging Face, models for high-capacity visual document retrieval. EVIE-8B reports state-of-the-art retrieval performance with 66.75 nDCG@10 on ViDoRe V3 and 4096D per-token multi-vector embeddings that preserve layout, typography, charts, and tables.
14 engagement1 sourceredditMon, Sep 7, 2026, 09:47 AM
Releases·Model Release
Tencent releases EVIE-Preview-4.5B, a multilingual visual document retrieval model
Tencent released EVIE-Preview-4.5B, a multilingual visual document retrieval model built on Qwen3.5-4B. It uses ColBERT-style late interaction with 128-dimensional multi-vector embeddings, cutting vector storage costs by 8× to 32×. The model achieves top-tier performance on ViDoRe benchmarks.
Tencent released EVIE-Preview-4.5B, a multilingual visual document retrieval model built on Qwen3.5-4B. It uses ColBERT-style late interaction with 128-dimensional multi-vector embeddings, cutting vector storage costs by 8× to 32×. The model achieves top-tier performance on ViDoRe benchmarks.
18 engagement2 sourcesreddit, rssMon, Aug 17, 2026, 12:24 PM
Releases·Model Release
NVIDIA releases Nemotron 3 Embed, a high-performance open embedding family
NVIDIA released the Nemotron 3 Embed family of open embedding models, claiming top performance on retrieval benchmarks. The models are available via NVIDIA NIM and a semantic search demo, aiming to improve RAG and agent workflows.
NVIDIA released the Nemotron 3 Embed family of open embedding models, claiming top performance on retrieval benchmarks. The models are available via NVIDIA NIM and a semantic search demo, aiming to improve RAG and agent workflows.
0 engagement1 sourcerssMon, Jul 27, 2026, 09:09 AM
Releases·Model Release
NVIDIA releases Nemotron-3-Embed 1B and 8B text embedding models
NVIDIA released Nemotron-3-Embed, a collection of open embedding models led by an 8B model that ranks #1 on the RTEB leaderboard, along with efficient 1B variants. The models support multilingual retrieval across 34 languages and are designed for production-scale RAG, agentic retrieval, and code retrieval. Amazon SageMaker AI now offers serverless fine-tuning for these models.
NVIDIA released Nemotron-3-Embed, a collection of open embedding models led by an 8B model that ranks #1 on the RTEB leaderboard, along with efficient 1B variants. The models support multilingual retrieval across 34 languages and are designed for production-scale RAG, agentic retrieval, and code retrieval. Amazon SageMaker AI now offers serverless fine-tuning for these models.
0304 engagement3 sourcesreddit, rssThu, Jul 16, 2026, 12:00 AM
Social·Community
User compares self-hosted vs managed embedding options for cutting reindex-heavy pipeline costs
A practitioner shares notes on reducing embedding costs after nightly re-embedding made their embedding spend exceed generation spend. They compare managed APIs (OpenAI, Cohere, Voyage) against self-hosted options like Hugging Face's TEI. Per-token pricing scales with corpus size and every reindex, so self-hosting can win for large, steady workloads.
A practitioner shares notes on reducing embedding costs after nightly re-embedding made their embedding spend exceed generation spend. They compare managed APIs (OpenAI, Cohere, Voyage) against self-hosted options like Hugging Face's TEI. Per-token pricing scales with corpus size and every reindex, so self-hosting can win for large, steady workloads.
10 engagement1 sourceredditTue, Sep 1, 2026, 11:19 AM