Model library / Embeddings

BGE-M3: GPU requirements and hosting

An embedding model for retrieval. It turns documents and queries into searchable representations; pair it with a text model when building retrieval-backed answers.

Sources reviewed 2026-09-30 · B3IQ engineering

Runtime profiles and requirements

Dense, sparse and multi-vector retrieval are documented upstream. The Ollama profile does not establish that all three interfaces are exposed, and GPU sizing needs to match the actual retrieval method.

ollama · embed-bge-m3

Catalog GPU threshold: Not specified
Configured context: 2,048 tokens
Catalog system RAM: 4 GB

Runtime: ollama
Artifact identifier: bge-m3
Precision: Not pinned in catalog
Configured concurrency: 1 request(s)
Profile inputs / outputs: text → embedding
Deployed artifact revision: Not verified

Catalog thresholds are configuration guidance, not measured peak memory. A positive GPU requirement is not recorded for every profile; unspecified is not zero. Context, concurrency and runtime overhead can increase memory use. Configured context is a profile setting, not the upstream maximum. Concurrency is configuration, not a load-test result.

Weight-only arithmetic and a planning estimate

Using the catalog's approximate 0.567B total parameters, 16-bit weights alone occupy about 1.13 GB (parameters × 2 bytes). The shared sizing helper rounds a 20% planning reserve to 1 GB.

This arithmetic does not describe the selected quantized artifact. It excludes a workload-specific KV-cache calculation and cannot guarantee fit at the configured context or concurrency. A mixture-of-experts model still stores its full weights.

See the assumptions →

Plan the machine.

Memory-based machine candidates depend on the current visible store configurations. Confirm runtime compatibility, GPU count, interconnect and workload before purchase. Pricing and availability are shown on the machine page.

Explore machines

Before deployment

Test the workload you need.

Judge retrieval on relevant documents found, not chat quality. Fix chunk size, batch size and retrieval method, and test the languages your corpus actually contains.

These are catalog profiles and planning figures. No B3IQ performance measurement or live capacity is claimed. Confirm the artifact, runtime, workload and machine before deployment.

Evidence available

Upstream identity / license: Source reviewed
Runtime settings: Catalog configuration
Memory fit: Guidance; not a measured peak
B3IQ performance: No published benchmark
Installation / live capacity: Confirm for your machine

Read our evidence standard →

Source and access.

Upstream checkpoint: BAAI/bge-m3 ↗
Upstream license metadata: MIT ↗
Access: No access gate reported by the upstream repository at review.
Reviewed source revision: 5617a9f61b028005a4858fdac845db406aefb181 ↗

The reviewed revision identifies the source used for this guide. It is not a claim that this revision is installed on a B3IQ machine. Review the publisher's current license and acceptable-use terms for your application.