SAVRN Model Hub · Comparisons
bge-small-en-v1.5 vs granite-embedding-small-english-r2
Bge-small-en-v1.5 has 33M parameters and granite-embedding-small-english-r2 has 48M parameters; bge-small-en-v1.5 is released under MIT License and granite-embedding-small-english-r2 under Apache License 2.0; at 16-bit, bge-small-en-v1.5 needs about 0.1 GB (1x MI300X from $1.85 an hour) and granite-embedding-small-english-r2 about 0.1 GB (1x MI300X from $1.85 an hour).
| Field | bge-small-en-v1.5 BAAI/bge-small-en-v1.5 | granite-embedding-small-english-r2 ibm-granite/granite-embedding-small-english-r2 |
|---|---|---|
| Publisher | Beijing Academy of Artificial Intelligence | IBM Granite |
| Task | Feature extraction | Feature extraction |
| Modality | Text | Text |
| Parameters, as reported | 33M parameters | 48M parameters |
| Architecture | BertModel | ModernBertModel |
| Library | sentence-transformers | sentence-transformers |
| Context length | 512 tokens | 8,192 tokens |
| Repository size | 401.1 MB | 194.3 MB |
| Artifact formats | safetensors, onnx, pytorch | safetensors, pytorch |
| License | mit | apache-2.0 |
| Access | Open weights, no gate | Open weights, no gate |
| Memory at 16-bit (weights and margin) | 0.1 GB | 0.1 GB |
| Cheapest GPUs at 16-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Memory at 4-bit (weights and margin) | 0 GB | 0 GB |
| Cheapest GPUs at 4-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Revision viewed | 5c38ec7c405e | 2ab6fa8ea2d6 |
| Downloads reported by the hub | 64.5M | 6.4M |
| Last observed | 2026-09-18 | 2026-09-18 |
An evaluation row appears only where at least two of these models report the same benchmark with the same stated configuration, metric, unit and setup. Different evaluators stay named in each cell. Values are shown as reported: no unit conversion, no ranking.
Other Reported Results
These results are listed for each model on its own, because the conditions needed to compare them are not stated or do not match. Two results that leave a condition blank are not assumed to share it.
bge-small-en-v1.5
| Benchmark | Conditions | Result | Reported by | Revision | Date |
|---|---|---|---|---|---|
| MTEB AmazonCounterfactualClassification (en) | Configuration enTask ClassificationMetric accuracyComparison conditions not established | 73.791 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB AmazonCounterfactualClassification (en) | Configuration enTask ClassificationMetric apComparison conditions not established | 37.2192 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB AmazonCounterfactualClassification (en) | Configuration enTask ClassificationMetric f1Comparison conditions not established | 68.0915 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB AmazonPolarityClassification | Configuration defaultTask ClassificationMetric accuracyComparison conditions not established | 92.7538 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB AmazonPolarityClassification | Configuration defaultTask ClassificationMetric apComparison conditions not established | 89.4677 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB AmazonPolarityClassification | Configuration defaultTask ClassificationMetric f1Comparison conditions not established | 92.7388 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB AmazonReviewsClassification (en) | Configuration enTask ClassificationMetric accuracyComparison conditions not established | 46.986 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB AmazonReviewsClassification (en) | Configuration enTask ClassificationMetric f1Comparison conditions not established | 46.5594 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric map_at_1Comparison conditions not established | 35.846 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric map_at_10Comparison conditions not established | 51.388 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric map_at_100Comparison conditions not established | 52.133 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric map_at_1000Comparison conditions not established | 52.141 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric map_at_3Comparison conditions not established | 47.037 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric map_at_5Comparison conditions not established | 49.579 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric mrr_at_1Comparison conditions not established | 36.558 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric mrr_at_10Comparison conditions not established | 51.658 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric mrr_at_100Comparison conditions not established | 52.402 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric mrr_at_1000Comparison conditions not established | 52.41 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric mrr_at_3Comparison conditions not established | 47.345 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric mrr_at_5Comparison conditions not established | 49.798 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric ndcg_at_1Comparison conditions not established | 35.846 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric ndcg_at_10Comparison conditions not established | 59.55 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric ndcg_at_100Comparison conditions not established | 62.596 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric ndcg_at_1000Comparison conditions not established | 62.759 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric ndcg_at_3Comparison conditions not established | 50.667 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric ndcg_at_5Comparison conditions not established | 55.228 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric precision_at_1Comparison conditions not established | 35.846 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric precision_at_10Comparison conditions not established | 8.542 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric precision_at_100Comparison conditions not established | 0.984 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric precision_at_1000Comparison conditions not established | 0.1 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric precision_at_3Comparison conditions not established | 20.389 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric precision_at_5Comparison conditions not established | 14.438 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric recall_at_1Comparison conditions not established | 35.846 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric recall_at_10Comparison conditions not established | 85.42 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric recall_at_100Comparison conditions not established | 98.435 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric recall_at_1000Comparison conditions not established | 99.644 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric recall_at_3Comparison conditions not established | 61.166 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna | Configuration defaultTask RetrievalMetric recall_at_5Comparison conditions not established | 72.191 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArxivClusteringP2P | Configuration defaultTask ClusteringMetric v_measureComparison conditions not established | 47.4028 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB ArxivClusteringS2S | Configuration defaultTask ClusteringMetric v_measureComparison conditions not established | 40.0155 | BAAI Publisher reported |
Evaluated revision not stated | — |
SAVRN's Notes on bge-small-en-v1.5
Twelve layers and 33M parameters come to 0.1 GB at 16-bit, with weights that round to zero at 8-bit and 4-bit, and that settles the hardware question. This embedder turns documents and queries into vectors for the retrieval-augmented LLM work its publisher focuses on, and it never justifies its own card: one MI300X, 192 GB, $1.85 an hour on demand is the cheapest setup we price, and this model would take a fraction of a percent. Put it on the GPU your generator already occupies.
MIT asks for one thing, that the copyright and permission notices stay attached to the weights, and allows commercial use, modification and redistribution in return. Hold the 512-token context and the en in the name up against your corpus; the publisher points to bge-m3 for more languages and inputs up to 8192 tokens. The files have not changed since 2024-02-22.
SAVRN's Notes on granite-embedding-small-english-r2
8,192 tokens of context on a 48M parameter embedding model change what you can index: a whole document section goes in as one unit. IBM Granite built this r2 release as a dense biencoder on ModernBERT, returning a 384-dimension vector per passage. Memory at 16-bit is 0.1 GB, and the cheapest listed host is one 192 GB MI300X at $1.85 an hour on demand, so plan on sharing that accelerator and paying for the slice of the hour you use.
Apache 2.0 carries an express patent grant, useful when embeddings ship inside a product; keep the license, notices and NOTICE file in the deployment and state significant changes. Before committing, read the training data, open relevance-pair datasets plus IBM's own collected and generated sets, against your data policy, and match the revision, released July 17, 2025 and last updated January 21, 2026, to the one you evaluated.
Questions
Which is larger, bge-small-en-v1.5 or granite-embedding-small-english-r2?
granite-embedding-small-english-r2 (48M parameters) is larger than bge-small-en-v1.5 (33M parameters), by the parameter counts their publishers report.
Which is cheaper to run, bge-small-en-v1.5 or granite-embedding-small-english-r2?
At 4-bit, bge-small-en-v1.5 fits on 1x MI300X from $1.85 an hour and granite-embedding-small-english-r2 on 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use bge-small-en-v1.5 commercially?
Yes. bge-small-en-v1.5 is released under MIT License. The MIT License is a short permissive license. It permits commercial use, modification and redistribution, provided the copyright notice and permission notice are included.
Can I use granite-embedding-small-english-r2 commercially?
Yes. granite-embedding-small-english-r2 is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.