mixedbread-ai/mxbai-colbert-large-v1
Primitive: /score · Score ·
BERT
The crispy rerank family from Mixedbread.
Overview
Hardware: — drives latency, throughput & cost
| Size | 335M params |
|---|---|
| Tasks | /encode · /score |
| License | apache-2.0 |
| Latency | 46 ms |
| Throughput | 4.0K tok/s |
| Cost | $0.056 /1M tok |
Cost is approximate — computed from list GPU prices; your actual price depends on the provider you deploy SIE with.
Scoring
| Inputs | text |
|---|---|
| Max sequence length | 512 |
Benchmarks
AskUbuntuDupQuestions
Duplicate question detection from AskUbuntu
Corpus: 6,743 Queries: 360
muvera
Quality
ndcg at 10 0.6343
map at 10 0.4758
mrr at 10 0.7104
default
Quality
ndcg at 10 0.6299
map at 10 0.4759
mrr at 10 0.7179
CMedQAv1-reranking
Chinese medical question answering reranking (v1)
Corpus: 100,000 Queries: 2,000
muvera
Quality
ndcg at 10 0.1654
map at 10 0.1166
mrr at 10 0.1711
default
Quality
ndcg at 10 0.2169
map at 10 0.1628
mrr at 10 0.2239
CQADupstackPhysicsRetrieval
Duplicate question retrieval from StackExchange Physics
Corpus: 38,314 Queries: 1,039
default_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.3503
mrr at 10 0.4030
ndcg at 10 0.4019
Performance L4 b1 c16
Corpus 30.0K tok/s
Corpus p50 68.7ms
Query 3.3K tok/s
Query p50 49.0ms
muvera_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.3368
mrr at 10 0.3886
ndcg at 10 0.3911
CosQA
Code search with natural language queries
Corpus: 6,267 Queries: 500
default_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.1957
mrr at 10 0.1874
ndcg at 10 0.2629
Performance L4 b1 c16
Corpus 12.5K tok/s
Corpus p50 62.5ms
Query 2.2K tok/s
Query p50 43.6ms
muvera_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.1842
mrr at 10 0.2013
ndcg at 10 0.2372
FiQA2018
Financial opinion mining and question answering
Corpus: 57,599 Queries: 648
default_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.2874
mrr at 10 0.4451
ndcg at 10 0.3629
Performance L4 b1 c16
Corpus 35.0K tok/s
Corpus p50 71.8ms
Query 4.2K tok/s
Query p50 44.7ms
muvera_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.2122
mrr at 10 0.3408
ndcg at 10 0.2787
LegalBenchConsumerContractsQA
Question answering on consumer contracts
Corpus: 153 Queries: 396
default_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.6024
mrr at 10 0.6003
ndcg at 10 0.6617
Performance L4 b1 c16
Corpus 77.3K tok/s
Corpus p50 101.4ms
Query 5.9K tok/s
Query p50 45.0ms
muvera_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.5411
mrr at 10 0.5403
ndcg at 10 0.6014
NFCorpus
Biomedical literature search from NutritionFacts.org
Corpus: 3,593 Queries: 323
default_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.1323
mrr at 10 0.5571
ndcg at 10 0.3434
Performance L4 b1 c16
Corpus 51.3K tok/s
Corpus p50 92.4ms
Query 1.8K tok/s
Query p50 44.1ms
muvera_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.1140
mrr at 10 0.5081
ndcg at 10 0.3127
SCIDOCS
Citation prediction, document classification, and recommendation for scientific papers
Corpus: 25,656 Queries: 1,000
default_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.0935
mrr at 10 0.2926
ndcg at 10 0.1603
Performance L4 b1 c16
Corpus 36.2K tok/s
Corpus p50 81.3ms
Query 3.8K tok/s
Query p50 46.5ms
muvera_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.0979
mrr at 10 0.2938
ndcg at 10 0.1683
SciFact
Scientific claim verification using research literature
Corpus: 5,183 Queries: 300
default_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.6803
mrr at 10 0.6897
ndcg at 10 0.7191
Performance L4 b1 c16
Corpus 46.8K tok/s
Corpus p50 89.5ms
Query 5.7K tok/s
Query p50 46.2ms
muvera_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.6189
mrr at 10 0.6280
ndcg at 10 0.6603
StackOverflowQA
Programming question answering from Stack Overflow
Corpus: 19,931 Queries: 1,994
default_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.6393
mrr at 10 0.6393
ndcg at 10 0.6728
Performance L4 b1 c16
Corpus 45.2K tok/s
Corpus p50 78.9ms
Query 69.9K tok/s
Query p50 64.7ms
muvera_candidates-k-50_candidates-model-Alibaba-NLP__gte-multilingual-base
Quality
map at 10 0.4550
mrr at 10 0.4550
ndcg at 10 0.5036
Compare (0)Compare →