Update README.md
Browse files
README.md
CHANGED
|
@@ -5,9 +5,7 @@ license_link: >-
|
|
| 5 |
https://developer.download.nvidia.com/licenses/nvidia-open-model-license-agreement-june-2024.pdf
|
| 6 |
---
|
| 7 |
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
**This has no support for inference, yet.** All I've done is move the weights out of NVIDIAs NeMo architecture so people smarter than me can get a headstart on making it work with other backends.
|
| 11 |
|
| 12 |
## Nemotron-4-340B-Instruct
|
| 13 |
|
|
|
|
| 5 |
https://developer.download.nvidia.com/licenses/nvidia-open-model-license-agreement-june-2024.pdf
|
| 6 |
---
|
| 7 |
|
| 8 |
+
Based on [nemotron3-8b](https://huggingface.co/thhaus/nemotron3-8b) and [Nemotron-4-340B-Instruct-SafeTensors](https://huggingface.co/failspy/Nemotron-4-340B-Instruct-SafeTensors) with quite a few changes to make compatible with vLLM, PR here: https://github.com/vllm-project/vllm/pull/6611
|
|
|
|
|
|
|
| 9 |
|
| 10 |
## Nemotron-4-340B-Instruct
|
| 11 |
|