SAVRN Model Hub · Comparisons
jina-embeddings-v3 vs Qwen3-Embedding-0.6B
Jina-embeddings-v3 has 572M parameters and Qwen3-Embedding-0.6B has 596M parameters; jina-embeddings-v3 is released under Creative Commons Attribution-NonCommercial 4.0 and Qwen3-Embedding-0.6B under Apache License 2.0; at 16-bit, jina-embeddings-v3 needs about 1.4 GB (1x MI300X from $1.85 an hour) and Qwen3-Embedding-0.6B about 1.4 GB (1x MI300X from $1.85 an hour).
| Field | jina-embeddings-v3 jinaai/jina-embeddings-v3 | Qwen3-Embedding-0.6B Qwen/Qwen3-Embedding-0.6B |
|---|---|---|
| Publisher | Jina AI | Qwen |
| Task | Feature extraction | Feature extraction |
| Modality | Text | Text |
| Parameters, as reported | 572M parameters | 596M parameters |
| Architecture | XLMRobertaModel | Qwen3ForCausalLM |
| Library | transformers | sentence-transformers |
| Context length | 8,194 tokens | 32,768 tokens |
| Repository size | 5.7 GB | 1.2 GB |
| Artifact formats | safetensors, onnx, pytorch | safetensors |
| License | cc-by-nc-4.0 | apache-2.0 |
| Access | Open weights, no gate | Open weights, no gate |
| Memory at 16-bit (weights and margin) | 1.4 GB | 1.4 GB |
| Cheapest GPUs at 16-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Memory at 4-bit (weights and margin) | 0.3 GB | 0.4 GB |
| Cheapest GPUs at 4-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Revision viewed | ab036b023d30 | 97b0c614be4d |
| Downloads reported by the hub | 2M | 8.5M |
| 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.
jina-embeddings-v3
| Benchmark | Conditions | Result | Reported by | Revision | Date |
|---|---|---|---|---|---|
| MTEB AFQMC (default) | Configuration defaultTask STSMetric cosine_pearsonComparison conditions not established | 41.7424 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB AFQMC (default) | Configuration defaultTask STSMetric cosine_spearmanComparison conditions not established | 43.4727 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB AFQMC (default) | Configuration defaultTask STSMetric euclidean_pearsonComparison conditions not established | 42.2446 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB AFQMC (default) | Configuration defaultTask STSMetric euclidean_spearmanComparison conditions not established | 43.5251 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB AFQMC (default) | Configuration defaultTask STSMetric main_scoreComparison conditions not established | 43.4727 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB AFQMC (default) | Configuration defaultTask STSMetric manhattan_pearsonComparison conditions not established | 42.0462 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB AFQMC (default) | Configuration defaultTask STSMetric manhattan_spearmanComparison conditions not established | 43.3088 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB AFQMC (default) | Configuration defaultTask STSMetric pearsonComparison conditions not established | 41.7424 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB AFQMC (default) | Configuration defaultTask STSMetric spearmanComparison conditions not established | 43.4727 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric main_scoreComparison conditions not established | 50.118 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric map_at_1Comparison conditions not established | 24.253 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric map_at_10Comparison conditions not established | 40.725 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric map_at_100Comparison conditions not established | 41.7 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric map_at_1000Comparison conditions not established | 41.707 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric map_at_20Comparison conditions not established | 41.468 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric map_at_3Comparison conditions not established | 35.467 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric map_at_5Comparison conditions not established | 38.291 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric mrr_at_1Comparison conditions not established | 24.7511 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric mrr_at_10Comparison conditions not established | 40.9106 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric mrr_at_100Comparison conditions not established | 41.8855 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric mrr_at_1000Comparison conditions not established | 41.893 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric mrr_at_20Comparison conditions not established | 41.6536 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric mrr_at_3Comparison conditions not established | 35.6567 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric mrr_at_5Comparison conditions not established | 38.4661 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric nauc_map_at_1000_diff1Comparison conditions not established | 7.503 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric nauc_map_at_1000_maxComparison conditions not established | -11.0304 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric nauc_map_at_1000_stdComparison conditions not established | -8.90279 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric nauc_map_at_100_diff1Comparison conditions not established | 7.5099 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric nauc_map_at_100_maxComparison conditions not established | -11.0236 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric nauc_map_at_100_stdComparison conditions not established | -8.89224 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric nauc_map_at_10_diff1Comparison conditions not established | 7.2437 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric nauc_map_at_10_maxComparison conditions not established | -10.81 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric nauc_map_at_10_stdComparison conditions not established | -8.98723 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric nauc_map_at_1_diff1Comparison conditions not established | 11.3718 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric nauc_map_at_1_maxComparison conditions not established | -13.3152 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric nauc_map_at_1_stdComparison conditions not established | -9.3982 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric nauc_map_at_20_diff1Comparison conditions not established | 7.47736 | jinaai Publisher reported |
Evaluated revision not stated | — |
| MTEB ArguAna-PL (default) | Configuration defaultTask RetrievalMetric nauc_map_at_20_maxComparison conditions not established | -10.9013 | jinaai Publisher reported |
Evaluated revision not stated | — |
| mteb/arguana | Task ArguAnaMetric ArguAnaSetup Obtained using MTEB v1.18.2Comparison conditions not established | 43.288 | Obtained using MTEB v1.18.2 Reported by a third party |
Evaluated revision not stated | 2026-03-05 |
| mteb/arguana | Task ArguAna_default_testMetric ArguAna_default_testSetup Obtained using MTEB v1.18.2Comparison conditions not established | 43.288 | Obtained using MTEB v1.18.2 Reported by a third party |
Evaluated revision not stated | 2026-03-05 |
SAVRN's Notes on jina-embeddings-v3
One point four gigabytes. That is jina-embeddings-v3 at 16-bit, 1.1 GB of it weights, dropping to 0.7 GB at 8-bit and 0.3 GB at 4-bit. It turns text into vectors for retrieval, clustering and classification across languages, with five LoRA adapters that switch the embedding to the task and an 8,194-token context. On the cheapest accelerator we price, one MI300X at $1.85 per hour with 192 GB, 572M parameters are a rounding error; it belongs beside a larger model on hardware you already run.
The license is the catch. CC BY-NC 4.0 permits sharing and adapting with credit for non-commercial purposes only, so a deployment that earns revenue needs separate permission from the rights holder; settle that in writing first. Then pick the published format that fits your stack, safetensors, ONNX or PyTorch, and plan disk for the 5.7 GB download even though a 16-bit load takes 1.4 GB.
SAVRN's Notes on Qwen3-Embedding-0.6B
Retrieval pipelines need an encoder that turns text into vectors, and this is the 0.6B of a series that also ships 4B and 8B. It needs 1.4 GB at 16-bit, 0.7 GB at 8-bit and 0.4 GB at 4-bit, so the cheapest setup, a 192 GB MI300X at $1.85 per hour, is a card it will share with the generator it feeds. The 32,768-token context matters more, since it sets how much of a document goes into one vector.
Apache 2.0 makes it a clean component to ship, with commercial use, modification and redistribution permitted, the license and NOTICE file kept and significant changes stated. It derives from Qwen3-0.6B-Base, and the architecture is Qwen3ForCausalLM served through sentence-transformers, so confirm your stack loads it that way. The Index lists no host price, so the comparison is your own accelerator time; arXiv:2506.05176 describes the method.
Questions
Which is larger, jina-embeddings-v3 or Qwen3-Embedding-0.6B?
Qwen3-Embedding-0.6B (596M parameters) is larger than jina-embeddings-v3 (572M parameters), by the parameter counts their publishers report.
Which is cheaper to run, jina-embeddings-v3 or Qwen3-Embedding-0.6B?
At 4-bit, jina-embeddings-v3 fits on 1x MI300X from $1.85 an hour and Qwen3-Embedding-0.6B on 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use jina-embeddings-v3 commercially?
Not without separate permission. jina-embeddings-v3 is released under Creative Commons Attribution-NonCommercial 4.0. CC BY-NC 4.0 permits sharing and adapting with credit for non-commercial purposes only. Commercial use needs separate permission from the rights holder.
Can I use Qwen3-Embedding-0.6B commercially?
Yes. Qwen3-Embedding-0.6B 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.