other
0.335B parameters
Commercial OK
Reviewed May 2026

BGE Large EN v1.5

BGE Large EN v1.5 is the 335M-parameter English flagship from BAAI's FlagEmbedding family, producing 1024-dim embeddings with a 512-token context window. Released in late 2023 under MIT license, it became the de facto MTEB benchmark anchor and remains one of the most-downloaded sentence-similarity models on the Hub.

License: mit·Context: 512 tokens
BLK · VERDICT

Our verdict

OP · Fredoline Eruo|VERIFIED MAY 29, 2026
unrated

The MTEB reference point that everyone benchmarks against. Still excellent for short-chunk English RAG but the 512-token ceiling shows its age — if you're starting fresh in 2026, nomic-v1.5 or gte-modernbert-base give you 8K context with comparable quality. Keep BGE if you're already paired with bge-reranker-v2-m3.

Overview

BGE Large EN v1.5 is the 335M-parameter English flagship from BAAI's FlagEmbedding family, producing 1024-dim embeddings with a 512-token context window. Released in late 2023 under MIT license, it became the de facto MTEB benchmark anchor and remains one of the most-downloaded sentence-similarity models on the Hub.

Strengths

  • 1024-dim dense embeddings with strong MTEB English score (~64.2)
  • MIT license — no commercial restrictions, training data sources documented
  • Massive ecosystem support: native in sentence-transformers, llama.cpp, ONNX, OpenVINO
  • v1.5 dropped the similarity-distribution issue of v1 — usable without prefix tuning

Weaknesses

  • Only 512-token context — chunking required for any real document
  • English-only; use bge-m3 or multilingual-e5 for non-English corpora
  • Architecture (BERT-large) is older — newer 8K-context models match quality at lower latency
  • No Matryoshka support — full 1024 dims always required at storage

Quantization variants

Each quantization trades model quality for file size and VRAM. Q4_K_M is the most popular starting point.

QuantizationFile sizeVRAM required
Q4_K_M0.2 GB1 GB

Get the model

HuggingFace

Original weights

huggingface.co/BAAI/bge-large-en-v1.5

Source repository — direct quantization required.

Hardware that runs this

Cards with enough VRAM for at least one quantization of BGE Large EN v1.5.

Compare alternatives

Models worth comparing

Same parameter band, plus what's one tier above and below — so you can decide what actually fits your hardware.

Step down
Smaller — faster, runs on weaker hardware
No verdicted models in the next tier down yet.

Frequently asked

What's the minimum VRAM to run BGE Large EN v1.5?

1GB of VRAM is enough to run BGE Large EN v1.5 at the Q4_K_M quantization (file size 0.2 GB). Higher-quality quantizations need more.

Can I use BGE Large EN v1.5 commercially?

Yes — BGE Large EN v1.5 ships under the mit, which permits commercial use. Always read the license text before deployment.

What's the context length of BGE Large EN v1.5?

BGE Large EN v1.5 supports a context window of 512 tokens (about 1K).

Source: huggingface.co/BAAI/bge-large-en-v1.5

Reviewed by RunLocalAI Editorial. See our editorial policy for how we research and verify model claims.

Related — keep moving

Before you buy

Verify BGE Large EN v1.5 runs on your specific hardware before committing money.