SAVRN Model Hub · Comparisons
Bangla-twoclass-Sentiment-Analyzer vs bge-reranker-base
Bangla-twoclass-Sentiment-Analyzer has 278M parameters and bge-reranker-base has 278M parameters; both are released under MIT License; at 16-bit, Bangla-twoclass-Sentiment-Analyzer needs about 0.7 GB (1x MI300X from $1.85 an hour) and bge-reranker-base about 0.7 GB (1x MI300X from $1.85 an hour).
| Field | Bangla-twoclass-Sentiment-Analyzer Arunavaonly/Bangla-twoclass-Sentiment-Analyzer | bge-reranker-base BAAI/bge-reranker-base |
|---|---|---|
| Publisher | Arunava Kar | Beijing Academy of Artificial Intelligence |
| Task | Text classification | Text classification |
| Modality | Text | Text |
| Parameters, as reported | 278M parameters | 278M parameters |
| Architecture | XLMRobertaForSequenceClassification | XLMRobertaForSequenceClassification |
| Library | transformers | sentence-transformers |
| Context length | 514 tokens | 514 tokens |
| Repository size | 2.2 GB | 3.4 GB |
| Artifact formats | safetensors, pytorch, tensorboard | safetensors, onnx, pytorch |
| License | mit | mit |
| Access | Open weights, no gate | Open weights, no gate |
| Memory at 16-bit (weights and margin) | 0.7 GB | 0.7 GB |
| Cheapest GPUs at 16-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Memory at 4-bit (weights and margin) | 0.2 GB | 0.2 GB |
| Cheapest GPUs at 4-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Revision viewed | b2b3ca5db5ea | 2cfc18c9415c |
| Downloads reported by the hub | 428.8k | 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-reranker-base
| Benchmark | Conditions | Result | Reported by | Revision | Date |
|---|---|---|---|---|---|
| MTEB CMedQAv1 | Configuration defaultTask RerankingMetric mapComparison conditions not established | 81.2721 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB CMedQAv1 | Configuration defaultTask RerankingMetric mrrComparison conditions not established | 84.1424 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB CMedQAv2 | Configuration defaultTask RerankingMetric mapComparison conditions not established | 84.1037 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB CMedQAv2 | Configuration defaultTask RerankingMetric mrrComparison conditions not established | 86.7938 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB MMarcoReranking | Configuration defaultTask RerankingMetric mapComparison conditions not established | 35.4601 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB MMarcoReranking | Configuration defaultTask RerankingMetric mrrComparison conditions not established | 34.6024 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB T2Reranking | Configuration defaultTask RerankingMetric mapComparison conditions not established | 67.2773 | BAAI Publisher reported |
Evaluated revision not stated | — |
| MTEB T2Reranking | Configuration defaultTask RerankingMetric mrrComparison conditions not established | 77.1315 | BAAI Publisher reported |
Evaluated revision not stated | — |
SAVRN's Notes on Bangla-twoclass-Sentiment-Analyzer
Two labels, one language. Bangla sentiment classification is the whole job for this 278M parameter model, fine-tuned from xlm-roberta-base for 1,800 steps at a batch size of 16, with the fine-tuning dataset left unnamed. At 16-bit the weights come to 0.6 GB and 0.7 GB of memory is needed. The cheapest configuration on the table is a single MI300X, 192 GB, $1.85 per hour on-demand, far more card than the work requires, so we would put it on shared capacity rather than a device of its own.
MIT terms allow commercial use, modification and redistribution with the notices intact, so the license will not block a deployment. What to check is the lineage and the data: the relation to FacebookAI/xlm-roberta-base is recorded, the training set is not, and no evaluation results are reported. The 514-token window suits short reviews and messages, and the files ship as safetensors and pytorch.
SAVRN's Notes on bge-reranker-base
Retrieval pipelines need a second pass that re-scores what the first-stage retriever returns, and that is this 278M-parameter classifier's job: it reads a query and a passage together and scores the match. Memory is not the decision. At 16-bit the weights are 0.6 GB and the run needs 0.7 GB, under one percent of the 192 GB MI300X we price at $1.85 an hour, so it never gets its own card; it rides beside whatever generation model lives there.
The MIT terms are short and permissive: commercial use, modification and redistribution are allowed provided the copyright and permission notices stay with the files. The 514-token context must hold query and passage together, so chunk to fit. And the publisher's summary says newer rerankers with larger inputs and more languages shipped March 18, 2024, so decide whether this version or a successor is what you standardize on.
Questions
Which is larger, Bangla-twoclass-Sentiment-Analyzer or bge-reranker-base?
Bangla-twoclass-Sentiment-Analyzer (278M parameters) is larger than bge-reranker-base (278M parameters), by the parameter counts their publishers report.
Which is cheaper to run, Bangla-twoclass-Sentiment-Analyzer or bge-reranker-base?
At 4-bit, Bangla-twoclass-Sentiment-Analyzer fits on 1x MI300X from $1.85 an hour and bge-reranker-base on 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use Bangla-twoclass-Sentiment-Analyzer commercially?
Yes. Bangla-twoclass-Sentiment-Analyzer 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 bge-reranker-base commercially?
Yes. bge-reranker-base 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.