Bge m3
7900°
BAAI | embedding model | MIT license | 4.3 GB as a torrent. Dense + sparse + multi-vector retrieval in one checkpoint. Full mirror of BAAI/bge-m3. For more details please refer to our github repo: https://github.com/FlagOpen/FlagEmbedding
Guides for this model
Best local LLM by use case (October 2026): the right open model for every job
What is The Hugging Bay? A mirror index for open models, and how to verify a download with its hashes
Open weights vs open source vs API-only: what you are actually allowed to do
How to Install Ollama and Run a Local AI (Windows, Mac, Linux)
What is a GGUF file? Quantized models explained (Q4, Q8, fp8, NF4)
Captain's notes
Notes from 2026-10-13
BGE-M3 is an embedding model from BAAI that packs dense retrieval, sparse retrieval, and multi-vector (ColBERT-style) retrieval into a single checkpoint. That triple capability in one model is the main reason to reach for it over a simpler embedder. It suits developers building search, RAG pipelines, or any system that needs flexible retrieval without juggling multiple models.
Known for
- Three retrieval modes — dense, sparse, multi-vector — in one checkpoint.
- MIT licence: permissive for commercial and research use.
- 4.3 GB download, available as a torrent mirror.
- Backed by BAAI, the same lab behind the broader BGE embedding family.
Watch out for
- This is an embedding model, not a generative one — it produces vectors, not text.
- Open-weights status is listed as unclear on this page; check the BAAI repository directly before building on it.
- No model card details are published here; consult the FlagEmbedding GitHub for full usage and performance details.
Best for
Prompts to try
Compare sentence similarity
I have two sentences: 'The transformer architecture revolutionised NLP' and 'Attention mechanisms changed how language models work.' Using BGE-M3 dense embeddings, explain how you would compute their cosine similarity and what a score of 0.85 versus 0.40 would tell me about their semantic relationship.
Design a RAG retrieval pipeline
I am building a retrieval-augmented generation system over a corpus of 50,000 technical documents. Walk me through how to use BGE-M3's three retrieval modes — dense, sparse, and multi-vector — together, when each mode helps most, and how to combine their scores at query time.
Pick the right retrieval mode
My search queries are short keyword lookups like 'invoice date field' but my documents are long legal contracts. Which of BGE-M3's retrieval modes — dense, sparse, or multi-vector — should I weight most heavily, and why? What are the trade-offs of each for this exact use case?
Compare with
- Claude Opus 5.5API-only model1,000,000 token context · $4.00/M input, $20.00/M output
- Claude Fable 5.1API-only model1,000,000 token context · $10.00/M input, $50.00/M output
- GPT-6 AstraAPI-only model1,050,000 token context · $10.00/M input, $50.00/M output
- Gemini 4 ArgonAPI-only model1,048,576 token context · $2.00/M input, $10.00/M output
More text models from baai
Questions people ask
- Can I run BGE-M3 locally?
- The model is available as a 4.3 GB torrent download and carries an MIT licence, so you can run it on your own hardware; check the FlagEmbedding GitHub for the recommended inference setup.
- What makes BGE-M3 different from other embedding models?
- Most embedding models support only dense retrieval; BGE-M3 adds sparse and multi-vector (ColBERT-style) retrieval in the same checkpoint, letting you pick or combine modes without loading separate models.
- Is this a chat or text-generation model?
- No — BGE-M3 is an embedding model that converts text into vectors for search and retrieval tasks; it does not generate text responses.
Captain's verdictIf you need all three retrieval modes without the overhead of separate models, BGE-M3 is the practical choice.
Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.
Ranking badge
Updated Oct 22, 2026
Made this model? Show this badge on your page.
Paste it into your Hugging Face card or Civitai description.
Also ranked: #76 Hottest Text / LLM text model
Not this creator?
Get your own badge
bge m3 Popularity Trend
Heat score, last 12 months
Heat score
7,900°
This week
n/a
Wasn't released/tracked
This month
n/a
Wasn't released/tracked
Past year
n/a
Wasn't released/tracked
7,000°
Floor
The scale runs from 0° to 9,999°. Everything below 7,000° is kept for models no source reports on yet, which carry a stand-in number there so lists stay complete. 7,000° is where measured data begins: a model with real data but almost no downloads sits at the floor, and the measurements fill the degrees above it. The floor says nothing about the model; only the degrees above it do, and they are what separates one model from another.
+900°
Torrent seeders
1 people keep serving the weights over torrents or hosted mirrors. Serving a file costs real bandwidth and says the weights are still wanted, so each seeder counts as 1,000 downloads in the download term.
Show detailed breakdown
Every number on this site can be checked. This model's heat score is 7,900°, and here it is built one ingredient at a time, with the running total after each: the floor, the degrees each source adds, the bonus for the kind of model and what the age of its weights takes away. The colors are the bands of the chart above. The glossary at the end defines every term.
- 1Start at the floor.Why 7,000° and not zero: the scale runs from 0° to 9,999°, and the part below 7,000° is reserved for models no source reports on yet, which carry a stand-in number there so the lists stay complete and never outrank a model with data. 7,000° is where measured data begins. A model with real data but almost no downloads sits here; everything that follows is measurement, and the room above the floor holds ten tenfold steps in downloads at 300° each. The floor is the same for every model, so it decides nothing; the degrees above it do.7,000°running total
- 2Add the downloads.1,000 download-equivalents in all, from Torrent seeders. Ten times the downloads adds the same 300°, so size matters but never runs away with the score.+900°7,900°running total
- =The heat score.Where this model stands on every list of this site. Nothing else adjusts it.7,900°
Glossary: the terms used here
- Heat score
- One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
- Floor
- The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
- Download-equivalents
- Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
- Log scale
- Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
- Likes
- Thumbs up and likes from every source, worth half a download each on the log scale.
- Adoption rate
- How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
- Measured and estimated
- Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
- Seeder
- Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
- AI Horde
- A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
- Generation sites
- Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
- OpenRouter tokens
- A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
- Full model (checkpoint)
- The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
- LoRA and embedding
- Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
- Family and weights
- The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
- Age
- Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
- Family boost
- A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
- Stand-in score
- A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
- New! (grace period)
- A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
- Trend
- How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
- Typical model
- The middle of the catalog: half the models with a figure grew faster over the period, half slower.
How this model's popularity has moved against the rest of the catalog: its growth in downloads, likes and usage over each period, measured against the typical model's growth over the same period, with its heat score at the time. Above the typical growth reads as a rise, below it as a fall. The past is reconstructed from today's totals, the model's first day and its measured download speed where two readings exist; a model that did not exist at the start of a period shows n/a.
Run Locally
How to install
Best local LLM by use case (October 2026): the right open model for every job
Oct 13, 2026
Torrents
baai/bge-m3
pytorch_model.bin
onnx/model.onnx_data
colbert_linear.pt
onnx/model.onnx
sparse_linear.pt
Torrents are shared peer to peer by their seeders; ModelTrends.ai hosts no files. Open the magnet link with a BitTorrent client and compare the finished file's checksum with the one on the listing.
Bge m3 is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models and check its heat score to see how it is trending.
Browse by Model Family
Text / LLM Models
434
Anima Models
528
Chroma Models
19
Fibo Models
2
Flux Models
1,496
Flux 2 / Klein Models
1,239
Flux 3 Models
3
Moss Models
3
Iris Models
1
MiniMax H3 Models
364
Hunyuan Models
204
Ideogram Models
20
Krea2 Models
717
Qwen Models
20
Qwen2 Models
47
Ltx2 Models
153