Update README.md
Browse files
README.md
CHANGED
|
@@ -13,7 +13,7 @@ base_model:
|
|
| 13 |
|
| 14 |
DMind-2 is a series of Web3 investment analysis language models designed to provide real-time, professional Web3 investment consulting services for individual investors and professional institutions. Standing on the shoulders of numerous open-source pioneers, we have successfully launched two model variants through innovative post-training techniques. Among these, Dmind-2-Pro demonstrates exceptional depth of understanding and analytical capabilities when addressing complex Web3 ecosystem challenges, delivering comprehensive insights that span from macroeconomic trends to microscopic on-chain behaviors.
|
| 15 |
|
| 16 |
-
## Model Variants(
|
| 17 |
- **Base Model**: GLM-4.5-Air
|
| 18 |
- **Parameters**: 107B
|
| 19 |
- **Training Duration**: 1 month of refined post-training
|
|
@@ -78,7 +78,7 @@ Safety alignment is another aspect we particularly emphasize. The Web3 investmen
|
|
| 78 |
|
| 79 |
## Performance Metrics
|
| 80 |
|
| 81 |
-
| Category | Benchmark (Metric) | DeepSeek-R1-0528 | gpt-oss-120b | Qwen3-235b-a22b | GLM-4.5-Air | **
|
| 82 |
| :--- | :--- | :--- | :--- | :--- | :--- | :--- |
|
| 83 |
| **General** | | | | | | |
|
| 84 |
| | MMLU-Pro (EM) | 84.0 | 90.0 | 80.6 | 81.4 | 83.1 |
|
|
|
|
| 13 |
|
| 14 |
DMind-2 is a series of Web3 investment analysis language models designed to provide real-time, professional Web3 investment consulting services for individual investors and professional institutions. Standing on the shoulders of numerous open-source pioneers, we have successfully launched two model variants through innovative post-training techniques. Among these, Dmind-2-Pro demonstrates exceptional depth of understanding and analytical capabilities when addressing complex Web3 ecosystem challenges, delivering comprehensive insights that span from macroeconomic trends to microscopic on-chain behaviors.
|
| 15 |
|
| 16 |
+
## Model Variants(DMind-2-107B)
|
| 17 |
- **Base Model**: GLM-4.5-Air
|
| 18 |
- **Parameters**: 107B
|
| 19 |
- **Training Duration**: 1 month of refined post-training
|
|
|
|
| 78 |
|
| 79 |
## Performance Metrics
|
| 80 |
|
| 81 |
+
| Category | Benchmark (Metric) | DeepSeek-R1-0528 | gpt-oss-120b | Qwen3-235b-a22b | GLM-4.5-Air | **DMind-2-107B(107B)** |
|
| 82 |
| :--- | :--- | :--- | :--- | :--- | :--- | :--- |
|
| 83 |
| **General** | | | | | | |
|
| 84 |
| | MMLU-Pro (EM) | 84.0 | 90.0 | 80.6 | 81.4 | 83.1 |
|