
BGE-M3
Model ranking
Not matched to the current verified ranking snapshot.
| Provider | Price | Provider model ID | Status | Context / max output | Source |
|---|---|---|---|---|---|
Cloudflare Workers AIbge-m3 | Standard Tariff not listedInput $0.0118USD / 1M tokens Updated 2026-11-22 · 15:43 UTCCloudflare Workers AI · StandardInput: $0.0118 Published Workers AI USD list tariffs. Free daily Neuron allowance is separate, not a zero unit price. Image tile/step components must be evaluated together; no flat image price is inferred. Checked 2026-11-22T15:43:15.988Z Price source ↗ | @cf/baai/bge-m3 | Catalog-listedConditionsPublic catalog listing; account eligibility and capacity can vary. Verified component tariffs are supplied by the existing pricing engine. | Not supplied | Official source ↗2026-10-20 |
BGE-M3 is BAAI’s multilingual retrieval model, designed around multiple retrieval representations and document lengths. Its hosted API should be evaluated for the specific representation it returns.
BGE-M3: suitable applications
Use it to build multilingual semantic retrieval over documents, knowledge bases and mixed-length passages.
Capabilities and technical details
The developer model supports dense, sparse and multi-vector retrieval approaches. This flexibility belongs to the model family; a hosted response may expose only part of that interface.
BGE-M3 accepts text and returns embeddings. The recorded input reference is 8,192 tokens, with any lower Cloudflare cap recorded separately. The recorded terms are MIT; the linked source explains their scope.
BGE-M3 limitations and selection
The Cloudflare catalog’s aggregate context figure differs from the model-card sequence length. It should not be interpreted as a verified per-document input allowance.
Choose the representation first, then assess retrieval relevance, chunk size and index cost. Do not mix vector spaces from unrelated embedding models in one similarity index.
BGE-M3 API pricing
BGE-M3 pricing is tied to the selected endpoint and its documented token components. Compare input, output and cached-input charges only where those components are actually quoted; an absent component is not an invented free rate.
Review the Cloudflare BGE-M3 endpoint reference and the BAAI model documentation. Compare the linked Cloudflare Workers AI provider profile with the xpu live model directory when selecting a deployment.
Specifications
- Catalog model ID
- baai/bge-m3
- Developer model ID / revision
- Not verified in cited sources
- Input modalities
- textCloudflare Workers AI endpoint; route-specific capabilities ↗
- Output modalities
- embeddingsCloudflare Workers AI endpoint; route-specific capabilities ↗
- Context / input token limit
- 8,192Developer model card / documentation ↗
- Maximum output tokens
- Not verified in cited sources
- Tool calling
- Not verified in cited sources
- Structured output
- Not verified in cited sources
- API parameters (source-specific)
- Not verified in cited sources
- Weight access
- Open weightsDeveloper model card / documentation ↗
- Weight license / access terms
- MITDeveloper weight license / hosted access terms ↗
- Total parameters (billions)
- Not verified in cited sources
- Active parameters (billions)
- Not verified in cited sources
- Embedding dimensions
- 1024Developer model card ↗
- Download size (GB)
- Depends on checkpoint and precision; no verified file size
- Release date
- Not verified in cited sources
- Notes
- Cloudflare deployment limits and transport options are recorded on its offering. Weight license is separate from hosted-service conditions. Context shown here is the publisher configuration; provider caps can be lower. BGE-M3 model card supports 8,192-token sequences; Cloudflare lists 60,000 context tokens without confirming a per-document cap, so the latter is not used as a sequence limit.
Catalog entries link to their specification and availability sources. This is a curated selection, not every model each provider offers. Best prices are the lowest collected USD tariffs, not a total-cost estimate. Token offers are selected by input price, with output from the same tariff; workload mix may change the cheapest provider. Listed variants may require batch processing, off-peak hours, or specific media settings. Price coverage varies by provider; collected tariffs include their source and observation time. Arena user-preference scores are checked automatically every six hours against the latest published Text / overall dataset. Source: Arena / LMArena; CC BY 4.0. Selection and catalog matching by XPU LIVE. They apply to the shown evaluation configuration; listed prices may use different settings. Models without verified scores are not assumed to be weaker.
