Update README.md for PolyKode_120B_alpha_0.0126
Browse files
README.md
CHANGED
|
@@ -5,6 +5,7 @@ license_link: https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia
|
|
| 5 |
base_model: nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4
|
| 6 |
tags:
|
| 7 |
- polykode
|
|
|
|
| 8 |
- code-modernization
|
| 9 |
- nemotron
|
| 10 |
- procogia
|
|
@@ -20,6 +21,15 @@ language:
|
|
| 20 |
|
| 21 |
This Hugging Face repository contains **product configuration and documentation** for running Polykode on the NVIDIA Nemotron base model. It does **not** redistribute modified model weights unless explicitly stated in a release.
|
| 22 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
## Base model
|
| 24 |
|
| 25 |
| | |
|
|
@@ -36,7 +46,7 @@ ProCogia's contributions (separate from NVIDIA weights):
|
|
| 36 |
|
| 37 |
- **Polykode** product name and branding
|
| 38 |
- System prompts for code modernization workflows
|
| 39 |
-
- LiteLLM / vLLM inference configuration (`--served-model-name
|
| 40 |
- Deployment references for ProCogia infrastructure
|
| 41 |
|
| 42 |
Under the NVIDIA Nemotron Open Model License, configuration and API aliases that bind by name to the base model are **not** Derivative Works of the Nemotron weights.
|
|
@@ -54,11 +64,23 @@ Download the NVIDIA base checkpoint, then serve with the Polykode API name:
|
|
| 54 |
```bash
|
| 55 |
python -m vllm.entrypoints.openai.api_server \
|
| 56 |
--model /path/to/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 \
|
| 57 |
-
--served-model-name
|
| 58 |
--host 0.0.0.0 \
|
| 59 |
--port 8000
|
| 60 |
```
|
| 61 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 62 |
Use the system prompt and LiteLLM config in this repository for Polykode behavior.
|
| 63 |
|
| 64 |
## Repository contents
|
|
@@ -67,7 +89,7 @@ Use the system prompt and LiteLLM config in this repository for Polykode behavio
|
|
| 67 |
|------|-------------|
|
| 68 |
| `NOTICE` | Attribution and legal notice (required for redistribution) |
|
| 69 |
| `LICENSE` | Licensing summary |
|
| 70 |
-
| `
|
| 71 |
| `NOTICE_WEIGHTS_ADDENDUM.txt` | Use only if fine-tuned weights are published |
|
| 72 |
|
| 73 |
## Legal
|
|
|
|
| 5 |
base_model: nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4
|
| 6 |
tags:
|
| 7 |
- polykode
|
| 8 |
+
- PolyKode_120B_alpha_0.0126
|
| 9 |
- code-modernization
|
| 10 |
- nemotron
|
| 11 |
- procogia
|
|
|
|
| 21 |
|
| 22 |
This Hugging Face repository contains **product configuration and documentation** for running Polykode on the NVIDIA Nemotron base model. It does **not** redistribute modified model weights unless explicitly stated in a release.
|
| 23 |
|
| 24 |
+
## Current release
|
| 25 |
+
|
| 26 |
+
| | |
|
| 27 |
+
|---|---|
|
| 28 |
+
| **API model name** | `PolyKode_120B_alpha_0.0126` |
|
| 29 |
+
| **Display name** | PolyKode 120B alpha 0.0126 |
|
| 30 |
+
| **Live endpoint** | https://dataegis.ai.procogia.com/v1 |
|
| 31 |
+
| **Open WebUI** | https://dataegis.ai.procogia.com |
|
| 32 |
+
|
| 33 |
## Base model
|
| 34 |
|
| 35 |
| | |
|
|
|
|
| 46 |
|
| 47 |
- **Polykode** product name and branding
|
| 48 |
- System prompts for code modernization workflows
|
| 49 |
+
- LiteLLM / vLLM inference configuration (`--served-model-name PolyKode_120B_alpha_0.0126`)
|
| 50 |
- Deployment references for ProCogia infrastructure
|
| 51 |
|
| 52 |
Under the NVIDIA Nemotron Open Model License, configuration and API aliases that bind by name to the base model are **not** Derivative Works of the Nemotron weights.
|
|
|
|
| 64 |
```bash
|
| 65 |
python -m vllm.entrypoints.openai.api_server \
|
| 66 |
--model /path/to/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 \
|
| 67 |
+
--served-model-name PolyKode_120B_alpha_0.0126 \
|
| 68 |
--host 0.0.0.0 \
|
| 69 |
--port 8000
|
| 70 |
```
|
| 71 |
|
| 72 |
+
## API example (OpenAI-compatible)
|
| 73 |
+
|
| 74 |
+
```bash
|
| 75 |
+
curl https://dataegis.ai.procogia.com/v1/chat/completions \
|
| 76 |
+
-H "Authorization: Bearer YOUR_API_KEY" \
|
| 77 |
+
-H "Content-Type: application/json" \
|
| 78 |
+
-d '{
|
| 79 |
+
"model": "PolyKode_120B_alpha_0.0126",
|
| 80 |
+
"messages": [{"role": "user", "content": "Translate this SAS snippet to Python."}]
|
| 81 |
+
}'
|
| 82 |
+
```
|
| 83 |
+
|
| 84 |
Use the system prompt and LiteLLM config in this repository for Polykode behavior.
|
| 85 |
|
| 86 |
## Repository contents
|
|
|
|
| 89 |
|------|-------------|
|
| 90 |
| `NOTICE` | Attribution and legal notice (required for redistribution) |
|
| 91 |
| `LICENSE` | Licensing summary |
|
| 92 |
+
| `litellm_config.polykode.yaml` | LiteLLM routing + Polykode system prompt |
|
| 93 |
| `NOTICE_WEIGHTS_ADDENDUM.txt` | Use only if fine-tuned weights are published |
|
| 94 |
|
| 95 |
## Legal
|