Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
pipeline_tag: text-generation
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
## Overview
|
| 7 |
+
|
| 8 |
+
**Prime Intellect** released **INTELLECT-2**, a 32 billion parameter large language model (LLM) trained through distributed reinforcement learning on globally donated GPU resources. Built on the **Qwen2** architecture and fine-tuned with the **prime-rl** framework, INTELLECT-2 demonstrates strong performance in math, coding, and logical reasoning.
|
| 9 |
+
|
| 10 |
+
This model leverages GRPO (Generalized Reinforcement Policy Optimization) over verifiable rewards, introducing asynchronous distributed RL training with enhanced stability techniques. While its primary focus was on verifiable mathematical and coding tasks, it remains compatible with general-purpose text generation tasks.
|
| 11 |
+
|
| 12 |
+
## Variants
|
| 13 |
+
|
| 14 |
+
### INTELLECT-2
|
| 15 |
+
|
| 16 |
+
| No | Variant | Branch | Cortex CLI command |
|
| 17 |
+
|----|----------------------------------------------------------------------------------|--------|-----------------------------------|
|
| 18 |
+
| 1 | [INTELLECT-2 (32B)](https://huggingface.co/cortexso/intellect-2/tree/32b-gguf-q2-k) | 32b | `cortex run intellect-2:32b` |
|
| 19 |
+
|
| 20 |
+
Each branch includes multiple GGUF quantized versions, optimized for various hardware configurations:
|
| 21 |
+
- **INTELLECT-2-32B:** q2_k, q3_k_l, q3_k_m, q3_k_s, q4_k_m, q4_k_s, q5_k_m, q5_k_s, q6_k, q8_0
|
| 22 |
+
|
| 23 |
+
## Use it with Jan (UI)
|
| 24 |
+
|
| 25 |
+
1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)
|
| 26 |
+
2. Use in Jan model Hub:
|
| 27 |
+
```bash
|
| 28 |
+
cortexso/intellect-2
|
| 29 |
+
```
|
| 30 |
+
|
| 31 |
+
## Use it with Cortex (CLI)
|
| 32 |
+
|
| 33 |
+
1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)
|
| 34 |
+
2. Run the model with command:
|
| 35 |
+
```bash
|
| 36 |
+
cortex run intellect-2
|
| 37 |
+
```
|
| 38 |
+
|
| 39 |
+
## Credits
|
| 40 |
+
|
| 41 |
+
- **Author:** Prime Intellect
|
| 42 |
+
- **Converter:** [Menlo Research](https://menlo.ai/)
|
| 43 |
+
- **Original License:** [Apache-2.0](https://choosealicense.com/licenses/apache-2.0/)
|
| 44 |
+
- **Paper:** [Intellect 2 Technical Report](https://storage.googleapis.com/public-technical-paper/INTELLECT_2_Technical_Report.pdf)
|