Add model_max_length (32768) to YAML and Model Details section
#3
by
dilawarm - opened
README.md
CHANGED
|
@@ -15,6 +15,7 @@ tags:
|
|
| 15 |
- medical
|
| 16 |
- multilingual
|
| 17 |
library_name: sentence-transformers
|
|
|
|
| 18 |
---
|
| 19 |
|
| 20 |
<img src="https://i.imgur.com/oxvhvQu.png"/>
|
|
@@ -33,6 +34,16 @@ The model supports flexible dimension projections (2560, 1280, 640, 320, 160, 80
|
|
| 33 |
|
| 34 |
This model is released under a non-commercial license. If you'd like a commercial license, please contact us at contact@zeroentropy.dev.
|
| 35 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 36 |
## How to Use
|
| 37 |
|
| 38 |
```python
|
|
@@ -69,17 +80,17 @@ The model can also be used through ZeroEntropy's [/models/embed](https://docs.ze
|
|
| 69 |
|
| 70 |
NDCG@10 scores between `zembed-1` and competing embedding models, averaged across public and private benchmarks per domain. Full per-benchmark breakdown [here](https://docs.google.com/spreadsheets/d/1qFXGZLMg6-O5tVLIJS3tpf5QNJxCHiiQtj35dZub4vY/edit?gid=0#gid=0).
|
| 71 |
|
| 72 |
-
| Domain
|
| 73 |
|------------------|----------------------|---------------|----------|-----------------|-------------------|---------------|--------------|--------|
|
| 74 |
-
| Finance
|
| 75 |
-
| Healthcare
|
| 76 |
-
| Legal
|
| 77 |
-
| Conversational
|
| 78 |
-
| Manufacturing
|
| 79 |
-
| Web Search
|
| 80 |
-
| Code
|
| 81 |
-
| STEM & Math
|
| 82 |
-
| Enterprise
|
| 83 |
-
| **Average**
|
| 84 |
-
|
| 85 |
-
|
|
|
|
| 15 |
- medical
|
| 16 |
- multilingual
|
| 17 |
library_name: sentence-transformers
|
| 18 |
+
model_max_length: 32768
|
| 19 |
---
|
| 20 |
|
| 21 |
<img src="https://i.imgur.com/oxvhvQu.png"/>
|
|
|
|
| 34 |
|
| 35 |
This model is released under a non-commercial license. If you'd like a commercial license, please contact us at contact@zeroentropy.dev.
|
| 36 |
|
| 37 |
+
## Model Details
|
| 38 |
+
|
| 39 |
+
| Property | Value |
|
| 40 |
+
|---|---|
|
| 41 |
+
| Parameters | 4B |
|
| 42 |
+
| Context Length | 32,768 tokens (32k) |
|
| 43 |
+
| Base Model | Qwen/Qwen3-4B |
|
| 44 |
+
| Embedding Dimensions | 2560, 1280, 640, 320, 160, 80, 40 |
|
| 45 |
+
| License | CC-BY-NC-4.0 |
|
| 46 |
+
|
| 47 |
## How to Use
|
| 48 |
|
| 49 |
```python
|
|
|
|
| 80 |
|
| 81 |
NDCG@10 scores between `zembed-1` and competing embedding models, averaged across public and private benchmarks per domain. Full per-benchmark breakdown [here](https://docs.google.com/spreadsheets/d/1qFXGZLMg6-O5tVLIJS3tpf5QNJxCHiiQtj35dZub4vY/edit?gid=0#gid=0).
|
| 82 |
|
| 83 |
+
| Domain | ZeroEntropy zembed-1 | voyage-4-nano | Qwen3 4B | Cohere Embed v4 | gemini-embed-001 | jina-v5-small | OpenAI Large | bge-m3 |
|
| 84 |
|------------------|----------------------|---------------|----------|-----------------|-------------------|---------------|--------------|--------|
|
| 85 |
+
| Finance | **0.4476** | 0.4227 | 0.3715 | 0.3670 | 0.3291 | 0.3576 | 0.3291 | 0.3085 |
|
| 86 |
+
| Healthcare | **0.6260** | 0.5356 | 0.5134 | 0.4750 | 0.5008 | 0.5132 | 0.5315 | 0.3620 |
|
| 87 |
+
| Legal | **0.6723** | 0.5957 | 0.5858 | 0.5894 | 0.6069 | 0.5716 | 0.5099 | 0.5207 |
|
| 88 |
+
| Conversational | **0.5385** | 0.4045 | 0.4034 | 0.4244 | 0.4247 | 0.4430 | 0.3988 | 0.3296 |
|
| 89 |
+
| Manufacturing | **0.5556** | 0.4857 | 0.4932 | 0.4919 | 0.4664 | 0.4725 | 0.4736 | 0.3736 |
|
| 90 |
+
| Web Search | 0.6165 | 0.5977 | 0.6914 | **0.7242** | 0.5881 | 0.6772 | 0.6750 | 0.6311 |
|
| 91 |
+
| Code | **0.6452** | 0.6415 | 0.6379 | 0.6277 | 0.6305 | 0.6354 | 0.6155 | 0.5584 |
|
| 92 |
+
| STEM & Math | **0.5283** | 0.5012 | 0.5219 | 0.4698 | 0.4840 | 0.3780 | 0.3905 | 0.3399 |
|
| 93 |
+
| Enterprise | **0.3750** | 0.3600 | 0.2935 | 0.2915 | 0.3224 | 0.3012 | 0.3307 | 0.2213 |
|
| 94 |
+
| **Average** | **0.5561** | **0.5050** | **0.5013** | **0.4957** | **0.4837** | **0.4833** | **0.4727** | **0.4050** |
|
| 95 |
+
|
| 96 |
+
Bar chart comparing zembed-1 NDCG@10 scores against competing embedding models across domains
|