yjoonjang commited on
Commit
53e42a4
·
verified ·
1 Parent(s): 815f7d4

Update evaluation: official Qwen3-Reranker and KaLM-Reranker-V1-Large-R2 added, seq-cls variants and ettin removed

Browse files

- Tables now use the official Qwen/Qwen3-Reranker-{0.6B,4B,8B} instead of the tomaarsen seq-cls conversions, add KaLM-Embedding/KaLM-Reranker-V1-Large-R2, and drop cross-encoder/ettin-reranker-1b-v1 (not multilingual).
- New figure; caption notes the non-uniform x-axis.
- PPS note: same ~550 sampled pairs per task for every model; KaLM uses sdpa.
- Model weights are unchanged.

Files changed (2) hide show
  1. README.md +15 -15
  2. assets/pps_vs_ndcg9.png +2 -2
README.md CHANGED
@@ -26,7 +26,7 @@ tags:
26
  ## Key Characteristics
27
 
28
  - **Strong Korean reranking:** 88.49 average nDCG@10 across nine Korean benchmarks, the highest average among the evaluated models with fewer than 4B parameters and within 1.1 points of Qwen3-Reranker-4B.
29
- - **Long-document reranking:** 81.63 nDCG@10 on MultiLongDocRetrieval (MLDR), ranking second only to Qwen3-Reranker-8B (82.20) among the evaluated models.
30
  - **Efficient:** 55.1 pairs per second averaged over nine benchmarks on an NVIDIA RTX A6000 48GB, about 2.3× Qwen3-Reranker-4B and 3.6× Qwen3-Reranker-8B under the measurement conditions described below.
31
 
32
  ## Model Overview
@@ -139,21 +139,22 @@ Evaluation was conducted using the [reranker-simple-benchmark](https://github.co
139
 
140
  <img src="./assets/pps_vs_ndcg9.png" width="700" alt="Reranking performance (mean nDCG@10) vs. throughput (mean PPS)">
141
 
142
- The figure plots mean nDCG@10 against mean PPS across the nine Korean benchmarks, including MLDR. Dot color marks model size. Throughput uses each model's own input limit and its best measured batch size, so it measures neither speed at equal token lengths nor end-to-end retrieval latency.
143
 
144
  ### Results — MTEB-ko-retrieval (9 subsets)
145
  <!-- **공식 9개 subset 을 모두 평가한 모델**의 9-subset mean NDCG@10, PPS -->
146
  | Model | Params | Mean NDCG@10 | Mean PPS |
147
  |---|---|---|---|
148
- | tomaarsen/Qwen3-Reranker-8B-seq-cls | 7.6B | 0.9004 | 15.1 |
149
- | tomaarsen/Qwen3-Reranker-4B-seq-cls | 4.0B | 0.8956 | 24.3 |
 
150
  | **nlpai-lab/KURE-Reranker-base** | 1.7B | 0.8849 | 55.1 |
151
  | **nlpai-lab/KURE-Reranker-nano** | 149M | 0.8808 | 473.7 |
152
  | zeroentropy/zerank-2-reranker | 4.0B | 0.8695 | 29.4 |
153
  | lightonai/LightOn-rerank-PW-4B | 4.5B | 0.8664 | 15.0 |
154
  | mixedbread-ai/mxbai-rerank-large-v2 | 1.5B | 0.8661 | 65.2 |
155
  | BAAI/bge-reranker-v2-m3 | 568M | 0.8586 | 404.1 |
156
- | tomaarsen/Qwen3-Reranker-0.6B-seq-cls | 596M | 0.8585 | 99.6 |
157
  | nvidia/llama-nemotron-rerank-1b-v2 | 1.2B | 0.8522 | 127.3 |
158
  | nlpai-lab/LAMAR-600m | 568M | 0.8406 | 408.8 |
159
  | dragonkue/bge-reranker-v2-m3-ko | 568M | 0.8263 | 401.5 |
@@ -161,9 +162,8 @@ The figure plots mean nDCG@10 against mean PPS across the nine Korean benchmarks
161
  | upskyy/ko-reranker-8k | 568M | 0.8085 | 404.0 |
162
  | Dongjin-kr/ko-reranker | 560M | 0.7950 | 482.8 |
163
  | telepix/PIXIE-Spell-Reranker-Preview-0.6B | 596M | 0.7806 | 99.6 |
164
- | cross-encoder/ettin-reranker-1b-v1 | 1.0B | 0.6901 | 49.7 |
165
 
166
- PPS is measured on a single NVIDIA RTX A6000 48GB per model and averaged over the nine benchmarks. Inputs are length-sorted and dynamically padded within each batch.
167
 
168
  `jinaai/jina-reranker-v3` and `jinaai/jina-reranker-v3.5` are listwise rerankers that cannot be compared under the same 8,192-token condition on MLDR, so they are excluded from the 9-subset table and their MLDR cells are left blank below.
169
 
@@ -171,8 +171,9 @@ PPS is measured on a single NVIDIA RTX A6000 48GB per model and averaged over th
171
 
172
  | Model | Params | Ko-StrategyQA | AutoRAGRetrieval | PublicHealthQA | BelebeleRetrieval | MIRACLRetrieval | MrTidyRetrieval | MultiLongDocRetrieval | SQuADKorV1Retrieval | LawIRKo |
173
  |---|---|---|---|---|---|---|---|---|---|---|
174
- | tomaarsen/Qwen3-Reranker-8B-seq-cls | 7.6B | 0.8679 | 0.9546 | 0.8893 | 0.9907 | 0.8490 | 0.8409 | 0.8220 | 0.9880 | 0.9014 |
175
- | tomaarsen/Qwen3-Reranker-4B-seq-cls | 4.0B | 0.8733 | 0.9707 | 0.8685 | 0.9906 | 0.8533 | 0.8321 | 0.8105 | 0.9861 | 0.8752 |
 
176
  | **nlpai-lab/KURE-Reranker-base** | 1.7B | 0.8548 | 0.9762 | 0.8716 | 0.9880 | 0.8371 | 0.7970 | 0.8163 | 0.9891 | 0.8343 |
177
  | **nlpai-lab/KURE-Reranker-nano** | 149M | 0.8582 | 0.9741 | 0.8475 | 0.9830 | 0.8400 | 0.8011 | 0.7966 | 0.9895 | 0.8368 |
178
  | jinaai/jina-reranker-v3.5 | 597M | 0.8539 | 0.9838 | 0.8094 | 0.9733 | 0.8565 | 0.8194 | — | 0.9887 | 0.8545 |
@@ -181,7 +182,7 @@ PPS is measured on a single NVIDIA RTX A6000 48GB per model and averaged over th
181
  | lightonai/LightOn-rerank-PW-4B | 4.5B | 0.8567 | 0.9321 | 0.8693 | 0.9882 | 0.8072 | 0.8091 | 0.7609 | 0.9803 | 0.7938 |
182
  | mixedbread-ai/mxbai-rerank-large-v2 | 1.5B | 0.8563 | 0.9531 | 0.8772 | 0.9778 | 0.7939 | 0.8771 | 0.6787 | 0.9681 | 0.8130 |
183
  | BAAI/bge-reranker-v2-m3 | 568M | 0.8487 | 0.9663 | 0.8475 | 0.9853 | 0.8129 | 0.8222 | 0.6690 | 0.9853 | 0.7906 |
184
- | tomaarsen/Qwen3-Reranker-0.6B-seq-cls | 596M | 0.8336 | 0.9308 | 0.8489 | 0.9779 | 0.8507 | 0.7359 | 0.7813 | 0.9808 | 0.7867 |
185
  | nvidia/llama-nemotron-rerank-1b-v2 | 1.2B | 0.8536 | 0.9480 | 0.8491 | 0.9883 | 0.8182 | 0.8026 | 0.6719 | 0.9858 | 0.7527 |
186
  | nlpai-lab/LAMAR-600m | 568M | 0.8461 | 0.9591 | 0.8225 | 0.9835 | 0.8214 | 0.7975 | 0.5679 | 0.9850 | 0.7822 |
187
  | dragonkue/bge-reranker-v2-m3-ko | 568M | 0.8232 | 0.9684 | 0.8708 | 0.9769 | 0.7573 | 0.6776 | 0.7061 | 0.9846 | 0.6721 |
@@ -189,14 +190,14 @@ PPS is measured on a single NVIDIA RTX A6000 48GB per model and averaged over th
189
  | upskyy/ko-reranker-8k | 568M | 0.8143 | 0.9230 | 0.8388 | 0.9291 | 0.7249 | 0.6998 | 0.5975 | 0.9718 | 0.7770 |
190
  | Dongjin-kr/ko-reranker | 560M | 0.8468 | 0.9014 | 0.7675 | 0.9759 | 0.8017 | 0.7772 | 0.3721 | 0.9785 | 0.7343 |
191
  | telepix/PIXIE-Spell-Reranker-Preview-0.6B | 596M | 0.8329 | 0.9794 | 0.8534 | 0.9777 | 0.8449 | 0.7650 | 0.1829 | 0.9850 | 0.6042 |
192
- | cross-encoder/ettin-reranker-1b-v1 | 1.0B | 0.6624 | 0.8901 | 0.7461 | 0.6914 | 0.7004 | 0.6659 | 0.3651 | 0.9590 | 0.5306 |
193
 
194
  ### Per-dataset PPS
195
 
196
  | Model | Params | Ko-StrategyQA | AutoRAGRetrieval | PublicHealthQA | BelebeleRetrieval | MIRACLRetrieval | MrTidyRetrieval | MultiLongDocRetrieval | SQuADKorV1Retrieval | LawIRKo |
197
  |---|---|---|---|---|---|---|---|---|---|---|
198
- | tomaarsen/Qwen3-Reranker-8B-seq-cls | 7.6B | 16.6 | 7.3 | 17.5 | 19.4 | 24.9 | 26.5 | 0.7 | 10.6 | 12.3 |
199
- | tomaarsen/Qwen3-Reranker-4B-seq-cls | 4.0B | 26.1 | 11.8 | 28.4 | 31.5 | 40.0 | 42.4 | 1.1 | 17.2 | 20.0 |
 
200
  | **nlpai-lab/KURE-Reranker-base** | 1.7B | 59.2 | 27.3 | 64.6 | 71.1 | 89.1 | 97.1 | 2.6 | 39.1 | 45.7 |
201
  | **nlpai-lab/KURE-Reranker-nano** | 149M | 475.4 | 233.2 | 569.7 | 630.9 | 765.8 | 855.2 | 16.5 | 330.8 | 386.0 |
202
  | jinaai/jina-reranker-v3.5 | 597M | 141.0 | 38.6 | 148.6 | 174.6 | 277.3 | 295.7 | — | 68.4 | 86.7 |
@@ -205,7 +206,7 @@ PPS is measured on a single NVIDIA RTX A6000 48GB per model and averaged over th
205
  | lightonai/LightOn-rerank-PW-4B | 4.5B | 16.2 | 7.3 | 18.6 | 19.5 | 23.4 | 26.1 | 0.7 | 10.7 | 12.4 |
206
  | mixedbread-ai/mxbai-rerank-large-v2 | 1.5B | 70.0 | 32.9 | 76.8 | 84.1 | 104.2 | 113.6 | 3.2 | 47.0 | 54.7 |
207
  | BAAI/bge-reranker-v2-m3 | 568M | 420.7 | 195.6 | 471.3 | 545.3 | 653.2 | 724.8 | 9.3 | 271.6 | 345.5 |
208
- | tomaarsen/Qwen3-Reranker-0.6B-seq-cls | 596M | 107.1 | 49.6 | 118.0 | 129.3 | 159.5 | 173.9 | 4.2 | 71.8 | 82.9 |
209
  | nvidia/llama-nemotron-rerank-1b-v2 | 1.2B | 133.7 | 54.4 | 146.9 | 163.0 | 220.0 | 243.1 | 3.9 | 84.9 | 95.9 |
210
  | nlpai-lab/LAMAR-600m | 568M | 418.0 | 196.5 | 480.5 | 557.2 | 656.7 | 742.5 | 9.2 | 275.0 | 343.8 |
211
  | dragonkue/bge-reranker-v2-m3-ko | 568M | 416.9 | 195.1 | 467.2 | 540.7 | 653.4 | 716.7 | 9.2 | 271.9 | 342.7 |
@@ -213,7 +214,6 @@ PPS is measured on a single NVIDIA RTX A6000 48GB per model and averaged over th
213
  | upskyy/ko-reranker-8k | 568M | 411.8 | 195.6 | 474.3 | 551.4 | 649.3 | 726.0 | 9.2 | 275.1 | 342.8 |
214
  | Dongjin-kr/ko-reranker | 560M | 520.8 | 258.0 | 510.1 | 554.8 | 747.6 | 765.2 | 272.7 | 334.4 | 381.2 |
215
  | telepix/PIXIE-Spell-Reranker-Preview-0.6B | 596M | 104.6 | 49.4 | 117.8 | 129.8 | 160.5 | 175.6 | 4.3 | 72.1 | 82.6 |
216
- | cross-encoder/ettin-reranker-1b-v1 | 1.0B | 52.4 | 19.9 | 51.4 | 65.6 | 92.7 | 97.2 | 3.8 | 31.2 | 33.4 |
217
 
218
  Batch size starts at 8 and doubles until an out-of-memory error. Samples are repeated to fill complete batches. After warmup, three full passes are timed with CUDA events. Throughput is the total number of processed pairs divided by the accumulated GPU forward time. The highest-throughput successful batch is reported. Tokenization, data loading, and CPU preprocessing are excluded. The batch-search approach is informed by the [Ettin reranker speed benchmark](https://huggingface.co/blog/ettin-reranker#speed).
219
 
 
26
  ## Key Characteristics
27
 
28
  - **Strong Korean reranking:** 88.49 average nDCG@10 across nine Korean benchmarks, the highest average among the evaluated models with fewer than 4B parameters and within 1.1 points of Qwen3-Reranker-4B.
29
+ - **Long-document reranking:** 81.63 nDCG@10 on MultiLongDocRetrieval (MLDR), ranking second only to Qwen3-Reranker-8B (82.55) among the evaluated models.
30
  - **Efficient:** 55.1 pairs per second averaged over nine benchmarks on an NVIDIA RTX A6000 48GB, about 2.3× Qwen3-Reranker-4B and 3.6× Qwen3-Reranker-8B under the measurement conditions described below.
31
 
32
  ## Model Overview
 
139
 
140
  <img src="./assets/pps_vs_ndcg9.png" width="700" alt="Reranking performance (mean nDCG@10) vs. throughput (mean PPS)">
141
 
142
+ The figure plots mean nDCG@10 against mean PPS across the nine Korean benchmarks, including MLDR. Dot color marks model size. Throughput uses each model's own input limit and its best measured batch size, so it measures neither speed at equal token lengths nor end-to-end retrieval latency. The x-axis is a non-uniform log scale that stretches 20–26 and compresses 26–50 pairs/s so that nearby points stay readable.
143
 
144
  ### Results — MTEB-ko-retrieval (9 subsets)
145
  <!-- **공식 9개 subset 을 모두 평가한 모델**의 9-subset mean NDCG@10, PPS -->
146
  | Model | Params | Mean NDCG@10 | Mean PPS |
147
  |---|---|---|---|
148
+ | Qwen/Qwen3-Reranker-8B | 8.2B | 0.9030 | 15.2 |
149
+ | KaLM-Embedding/KaLM-Reranker-V1-Large-R2 | 7.5B | 0.8960 | 25.3 |
150
+ | Qwen/Qwen3-Reranker-4B | 4.0B | 0.8957 | 24.3 |
151
  | **nlpai-lab/KURE-Reranker-base** | 1.7B | 0.8849 | 55.1 |
152
  | **nlpai-lab/KURE-Reranker-nano** | 149M | 0.8808 | 473.7 |
153
  | zeroentropy/zerank-2-reranker | 4.0B | 0.8695 | 29.4 |
154
  | lightonai/LightOn-rerank-PW-4B | 4.5B | 0.8664 | 15.0 |
155
  | mixedbread-ai/mxbai-rerank-large-v2 | 1.5B | 0.8661 | 65.2 |
156
  | BAAI/bge-reranker-v2-m3 | 568M | 0.8586 | 404.1 |
157
+ | Qwen/Qwen3-Reranker-0.6B | 596M | 0.8577 | 100.2 |
158
  | nvidia/llama-nemotron-rerank-1b-v2 | 1.2B | 0.8522 | 127.3 |
159
  | nlpai-lab/LAMAR-600m | 568M | 0.8406 | 408.8 |
160
  | dragonkue/bge-reranker-v2-m3-ko | 568M | 0.8263 | 401.5 |
 
162
  | upskyy/ko-reranker-8k | 568M | 0.8085 | 404.0 |
163
  | Dongjin-kr/ko-reranker | 560M | 0.7950 | 482.8 |
164
  | telepix/PIXIE-Spell-Reranker-Preview-0.6B | 596M | 0.7806 | 99.6 |
 
165
 
166
+ PPS is measured on a single NVIDIA RTX A6000 48GB per model and averaged over the nine benchmarks. For each benchmark, every model is timed on the same ~550 sampled query–document pairs (whole queries, fixed seed). Inputs are length-sorted and dynamically padded within each batch. Models run in bf16 with flash_attention_2, except `KaLM-Embedding/KaLM-Reranker-V1-Large-R2`, which uses sdpa because transformers does not support flash_attention_2 for T5Gemma2.
167
 
168
  `jinaai/jina-reranker-v3` and `jinaai/jina-reranker-v3.5` are listwise rerankers that cannot be compared under the same 8,192-token condition on MLDR, so they are excluded from the 9-subset table and their MLDR cells are left blank below.
169
 
 
171
 
172
  | Model | Params | Ko-StrategyQA | AutoRAGRetrieval | PublicHealthQA | BelebeleRetrieval | MIRACLRetrieval | MrTidyRetrieval | MultiLongDocRetrieval | SQuADKorV1Retrieval | LawIRKo |
173
  |---|---|---|---|---|---|---|---|---|---|---|
174
+ | Qwen/Qwen3-Reranker-8B | 8.2B | 0.8720 | 0.9653 | 0.8899 | 0.9908 | 0.8513 | 0.8416 | 0.8255 | 0.9897 | 0.9013 |
175
+ | KaLM-Embedding/KaLM-Reranker-V1-Large-R2 | 7.5B | 0.8796 | 0.9415 | 0.8956 | 0.9936 | 0.8382 | 0.8500 | 0.7970 | 0.9863 | 0.8824 |
176
+ | Qwen/Qwen3-Reranker-4B | 4.0B | 0.8733 | 0.9630 | 0.8702 | 0.9904 | 0.8559 | 0.8298 | 0.8161 | 0.9861 | 0.8766 |
177
  | **nlpai-lab/KURE-Reranker-base** | 1.7B | 0.8548 | 0.9762 | 0.8716 | 0.9880 | 0.8371 | 0.7970 | 0.8163 | 0.9891 | 0.8343 |
178
  | **nlpai-lab/KURE-Reranker-nano** | 149M | 0.8582 | 0.9741 | 0.8475 | 0.9830 | 0.8400 | 0.8011 | 0.7966 | 0.9895 | 0.8368 |
179
  | jinaai/jina-reranker-v3.5 | 597M | 0.8539 | 0.9838 | 0.8094 | 0.9733 | 0.8565 | 0.8194 | — | 0.9887 | 0.8545 |
 
182
  | lightonai/LightOn-rerank-PW-4B | 4.5B | 0.8567 | 0.9321 | 0.8693 | 0.9882 | 0.8072 | 0.8091 | 0.7609 | 0.9803 | 0.7938 |
183
  | mixedbread-ai/mxbai-rerank-large-v2 | 1.5B | 0.8563 | 0.9531 | 0.8772 | 0.9778 | 0.7939 | 0.8771 | 0.6787 | 0.9681 | 0.8130 |
184
  | BAAI/bge-reranker-v2-m3 | 568M | 0.8487 | 0.9663 | 0.8475 | 0.9853 | 0.8129 | 0.8222 | 0.6690 | 0.9853 | 0.7906 |
185
+ | Qwen/Qwen3-Reranker-0.6B | 596M | 0.8336 | 0.9321 | 0.8477 | 0.9779 | 0.8470 | 0.7325 | 0.7819 | 0.9804 | 0.7865 |
186
  | nvidia/llama-nemotron-rerank-1b-v2 | 1.2B | 0.8536 | 0.9480 | 0.8491 | 0.9883 | 0.8182 | 0.8026 | 0.6719 | 0.9858 | 0.7527 |
187
  | nlpai-lab/LAMAR-600m | 568M | 0.8461 | 0.9591 | 0.8225 | 0.9835 | 0.8214 | 0.7975 | 0.5679 | 0.9850 | 0.7822 |
188
  | dragonkue/bge-reranker-v2-m3-ko | 568M | 0.8232 | 0.9684 | 0.8708 | 0.9769 | 0.7573 | 0.6776 | 0.7061 | 0.9846 | 0.6721 |
 
190
  | upskyy/ko-reranker-8k | 568M | 0.8143 | 0.9230 | 0.8388 | 0.9291 | 0.7249 | 0.6998 | 0.5975 | 0.9718 | 0.7770 |
191
  | Dongjin-kr/ko-reranker | 560M | 0.8468 | 0.9014 | 0.7675 | 0.9759 | 0.8017 | 0.7772 | 0.3721 | 0.9785 | 0.7343 |
192
  | telepix/PIXIE-Spell-Reranker-Preview-0.6B | 596M | 0.8329 | 0.9794 | 0.8534 | 0.9777 | 0.8449 | 0.7650 | 0.1829 | 0.9850 | 0.6042 |
 
193
 
194
  ### Per-dataset PPS
195
 
196
  | Model | Params | Ko-StrategyQA | AutoRAGRetrieval | PublicHealthQA | BelebeleRetrieval | MIRACLRetrieval | MrTidyRetrieval | MultiLongDocRetrieval | SQuADKorV1Retrieval | LawIRKo |
197
  |---|---|---|---|---|---|---|---|---|---|---|
198
+ | Qwen/Qwen3-Reranker-8B | 8.2B | 16.5 | 7.4 | 17.6 | 19.4 | 25.0 | 26.6 | 0.7 | 10.8 | 12.4 |
199
+ | KaLM-Embedding/KaLM-Reranker-V1-Large-R2 | 7.5B | 28.6 | 12.4 | 28.5 | 33.7 | 39.6 | 43.8 | 0.8 | 18.7 | 21.4 |
200
+ | Qwen/Qwen3-Reranker-4B | 4.0B | 26.2 | 11.8 | 28.2 | 31.1 | 40.1 | 42.8 | 1.1 | 17.1 | 19.9 |
201
  | **nlpai-lab/KURE-Reranker-base** | 1.7B | 59.2 | 27.3 | 64.6 | 71.1 | 89.1 | 97.1 | 2.6 | 39.1 | 45.7 |
202
  | **nlpai-lab/KURE-Reranker-nano** | 149M | 475.4 | 233.2 | 569.7 | 630.9 | 765.8 | 855.2 | 16.5 | 330.8 | 386.0 |
203
  | jinaai/jina-reranker-v3.5 | 597M | 141.0 | 38.6 | 148.6 | 174.6 | 277.3 | 295.7 | — | 68.4 | 86.7 |
 
206
  | lightonai/LightOn-rerank-PW-4B | 4.5B | 16.2 | 7.3 | 18.6 | 19.5 | 23.4 | 26.1 | 0.7 | 10.7 | 12.4 |
207
  | mixedbread-ai/mxbai-rerank-large-v2 | 1.5B | 70.0 | 32.9 | 76.8 | 84.1 | 104.2 | 113.6 | 3.2 | 47.0 | 54.7 |
208
  | BAAI/bge-reranker-v2-m3 | 568M | 420.7 | 195.6 | 471.3 | 545.3 | 653.2 | 724.8 | 9.3 | 271.6 | 345.5 |
209
+ | Qwen/Qwen3-Reranker-0.6B | 596M | 105.9 | 49.4 | 117.9 | 130.3 | 162.6 | 177.4 | 4.3 | 70.8 | 83.6 |
210
  | nvidia/llama-nemotron-rerank-1b-v2 | 1.2B | 133.7 | 54.4 | 146.9 | 163.0 | 220.0 | 243.1 | 3.9 | 84.9 | 95.9 |
211
  | nlpai-lab/LAMAR-600m | 568M | 418.0 | 196.5 | 480.5 | 557.2 | 656.7 | 742.5 | 9.2 | 275.0 | 343.8 |
212
  | dragonkue/bge-reranker-v2-m3-ko | 568M | 416.9 | 195.1 | 467.2 | 540.7 | 653.4 | 716.7 | 9.2 | 271.9 | 342.7 |
 
214
  | upskyy/ko-reranker-8k | 568M | 411.8 | 195.6 | 474.3 | 551.4 | 649.3 | 726.0 | 9.2 | 275.1 | 342.8 |
215
  | Dongjin-kr/ko-reranker | 560M | 520.8 | 258.0 | 510.1 | 554.8 | 747.6 | 765.2 | 272.7 | 334.4 | 381.2 |
216
  | telepix/PIXIE-Spell-Reranker-Preview-0.6B | 596M | 104.6 | 49.4 | 117.8 | 129.8 | 160.5 | 175.6 | 4.3 | 72.1 | 82.6 |
 
217
 
218
  Batch size starts at 8 and doubles until an out-of-memory error. Samples are repeated to fill complete batches. After warmup, three full passes are timed with CUDA events. Throughput is the total number of processed pairs divided by the accumulated GPU forward time. The highest-throughput successful batch is reported. Tokenization, data loading, and CPU preprocessing are excluded. The batch-search approach is informed by the [Ettin reranker speed benchmark](https://huggingface.co/blog/ettin-reranker#speed).
219
 
assets/pps_vs_ndcg9.png CHANGED

Git LFS Details

  • SHA256: 5b10c89e253670c6d812bd652cee8c9b9591983ede58a9d7ce7472fde6653c95
  • Pointer size: 131 Bytes
  • Size of remote file: 113 kB

Git LFS Details

  • SHA256: 8ecd1b0fb6e76331af76672d4ccf542f3395f0e57ef8ce85aae3abb36af76b47
  • Pointer size: 131 Bytes
  • Size of remote file: 144 kB