SAVRN Model Hub · Comparisons
indonesian-roberta-base-posp-tagger vs span-marker-bert-base-uncased-acronyms
Indonesian-roberta-base-posp-tagger has 124M parameters and span-marker-bert-base-uncased-acronyms has 109M parameters; indonesian-roberta-base-posp-tagger is released under MIT License and span-marker-bert-base-uncased-acronyms under Apache License 2.0; at 16-bit, indonesian-roberta-base-posp-tagger needs about 0.3 GB (1x MI300X from $1.85 an hour) and span-marker-bert-base-uncased-acronyms about 0.3 GB (1x MI300X from $1.85 an hour).
| Field | indonesian-roberta-base-posp-tagger w11wo/indonesian-roberta-base-posp-tagger | span-marker-bert-base-uncased-acronyms tomaarsen/span-marker-bert-base-uncased-acronyms |
|---|---|---|
| Publisher | Wilson Wongso | Tom Aarsen |
| Task | Token classification | Token classification |
| Modality | Text | Text |
| Parameters, as reported | 124M parameters | 109M parameters |
| Architecture | RobertaForTokenClassification | SpanMarkerModel |
| Library | transformers | span-marker |
| Context length | 514 tokens | Not stated |
| Repository size | 1.5 GB | 877.0 MB |
| Artifact formats | safetensors, pytorch, tensorboard, tf | safetensors, pytorch, tensorboard |
| License | mit | apache-2.0 |
| Access | Open weights, no gate | Open weights, no gate |
| Memory at 16-bit (weights and margin) | 0.3 GB | 0.3 GB |
| Cheapest GPUs at 16-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Memory at 4-bit (weights and margin) | 0.1 GB | 0.1 GB |
| Cheapest GPUs at 4-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Revision viewed | 506779636093 | d547e2eba6f6 |
| Downloads reported by the hub | 2.6M | 239.9k |
| 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.
indonesian-roberta-base-posp-tagger
| Benchmark | Conditions | Result | Reported by | Revision | Date |
|---|---|---|---|---|---|
| indonlu | Configuration pospTask Token ClassificationMetric AccuracyComparison conditions not established | 0.96251 | w11wo Publisher reported |
Evaluated revision not stated | — |
| indonlu | Configuration pospTask Token ClassificationMetric F1Comparison conditions not established | 0.96251 | w11wo Publisher reported |
Evaluated revision not stated | — |
| indonlu | Configuration pospTask Token ClassificationMetric PrecisionComparison conditions not established | 0.96251 | w11wo Publisher reported |
Evaluated revision not stated | — |
| indonlu | Configuration pospTask Token ClassificationMetric RecallComparison conditions not established | 0.96251 | w11wo Publisher reported |
Evaluated revision not stated | — |
span-marker-bert-base-uncased-acronyms
| Benchmark | Conditions | Result | Reported by | Revision | Date |
|---|---|---|---|---|---|
| Acronym Identification | Task Named Entity RecognitionMetric F1Comparison conditions not established | 0.919893 | tomaarsen Publisher reported |
Evaluated revision not stated | — |
| Acronym Identification | Task Named Entity RecognitionMetric PrecisionComparison conditions not established | 0.93394 | tomaarsen Publisher reported |
Evaluated revision not stated | — |
| Acronym Identification | Task Named Entity RecognitionMetric RecallComparison conditions not established | 0.906263 | tomaarsen Publisher reported |
Evaluated revision not stated | — |
Questions
Which is larger, indonesian-roberta-base-posp-tagger or span-marker-bert-base-uncased-acronyms?
indonesian-roberta-base-posp-tagger (124M parameters) is larger than span-marker-bert-base-uncased-acronyms (109M parameters), by the parameter counts their publishers report.
Which is cheaper to run, indonesian-roberta-base-posp-tagger or span-marker-bert-base-uncased-acronyms?
At 4-bit, indonesian-roberta-base-posp-tagger fits on 1x MI300X from $1.85 an hour and span-marker-bert-base-uncased-acronyms on 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use indonesian-roberta-base-posp-tagger commercially?
Yes. indonesian-roberta-base-posp-tagger 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 span-marker-bert-base-uncased-acronyms commercially?
Yes. span-marker-bert-base-uncased-acronyms 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.