bge-large-en-v1.5
- Open weights
Prices
0 offerings · prices as listedNo priced offering is listed for this model yet.
Run locally
Weights: BAAI/bge-large-en-v1.5- Parameters 0.3B
- Dense
- Native precision FP32
- Layers 24
- KV cache per token 96 KB
- Config positions 512 (config.json max_position_embeddings)
- Needs
- ≈ 1.8 GB (weights 1.2 (maker’s file) + KV cache 0.0 + overhead 0.5)
- Usable
- 21.6 GB (90% of its memory)
- Generation speed
- 362–544 tok/s
- Longest context
- 512 (the model’s limit)
- First token
- about 0.01–0.02 s for a 2K prompt
| Precision | Weights | Needs at 512 | On your hardware | Longest context |
|---|---|---|---|---|
| FP32 (native)maker’s file | 1.2 GB | ≈ 1.8 | ●Runs well 362–544 tok/s | 512 |
| Q8_0 | 0.3 GB | ≈ 0.9 | ●Runs well 1240–1860 tok/s | 512 |
| FP8 | 0.3 GB | ≈ 0.9 | ●Runs well 1307–1961 tok/s | 512 |
| Q6_K | 0.3 GB | ≈ 0.8 | ●Runs well 1548–2322 tok/s | 512 |
| Q5_K_M | 0.2 GB | ≈ 0.8 | ●Runs well 1746–2619 tok/s | 512 |
| Q5_0 | 0.2 GB | ≈ 0.8 | ●Runs well 1795–2693 tok/s | 512 |
| Q4_K_M | 0.2 GB | ≈ 0.7 | ●Runs well 1985–2977 tok/s | 512 |
| Q4_0 | 0.2 GB | ≈ 0.7 | ●Runs well 2110–3165 tok/s | 512 |
| IQ4_XS | 0.2 GB | ≈ 0.7 | ●Runs well 2207–3310 tok/s | 512 |
| Q3_K_Mquality drops | 0.2 GB | ≈ 0.7 | ●Runs well 2354–3531 tok/s | 512 |
| IQ3_XXSquality drops | 0.1 GB | ≈ 0.7 | ●Runs well 2823–4235 tok/s | 512 |
| Q2_Kquality drops | 0.1 GB | ≈ 0.7 | ●Runs well 2863–4295 tok/s | 512 |
| IQ2_XXSquality drops | 0.1 GB | ≈ 0.6 | ●Runs well 3689–5533 tok/s | 512 |
Click a row to choose a precision. “Maker’s file” sizes are the maker’s own files; others are parameters × bits per weight.
Compared with other hardware
Each at the highest precision it runs. Click one for details.
Smallest that runs it (Q4_K_M, 8K): GPU GeForce RTX 3080; unified memory Apple M5 · 16 GB.
Run commands
vllm serve BAAI/bge-large-en-v1.5 --max-model-len 512
Memory to fine-tune
Batch 1, 2,048-token sequences, gradient checkpointing; LoRA and QLoRA train rank-16 adapters on the attention projections.
Estimates; facts from the maker’s Hugging Face repositories, as of 2026-11-06. How it’s estimated
Cited scores
From third-party leaderboards, not measured by usNo cited leaderboard covers embedding models yet. MTEB’s results are open (CC0); we will cite them per benchmark, never as a combined score.
Changes
All changes →No changes recorded yet.
Sources
- BAAI/bge-large-en-v1.5 model card (opens in a new tab) huggingface.co · checked Nov 1, 2026
- BAAI/bge-large-en-v1.5 model card (opens in a new tab) huggingface.co · checked Nov 5, 2026
- BAAI/bge-large-en-v1.5 config.json (opens in a new tab) huggingface.co · checked Nov 2, 2026
- BAAI/bge-large-en-v1.5 files (opens in a new tab) huggingface.co · checked Nov 2, 2026
- BGE-M3 ([paper](https://arxiv.org/pdf/2402.03216.pdf), [code](https://github.com/FlagOpen/FlagEmbedding/tree/master/FlagEmbedding/BGEM3)) (opens in a new tab) huggingface.co · checked Nov 2, 2026
Some values on this page are still waiting to be checked against official pages: a filled dot is verified, a hollow one is not yet (with or without a source on file). How we verify data
BAAI · China · baai.ac.cn
Related models
Snapshots, the same family and other models from this provider