PrimeSim commited on
Commit ·
2b213fa
1
Parent(s): b8000e9
Point at the repo demos for the posttrain checkpoint
Browse files
README.md
CHANGED
|
@@ -34,6 +34,9 @@ it leads four of five WorldSimProbe suites and posts the best overall EWMBench s
|
|
| 34 |
| `primesim_base.safetensors` | the midtrained model -- start here to fine-tune |
|
| 35 |
| `primesim_posttrain_g2.safetensors` | stage 4, causal few-step model -- use for inference |
|
| 36 |
|
|
|
|
|
|
|
|
|
|
| 37 |
## Requirements
|
| 38 |
|
| 39 |
**These weights are not self-contained.** The text encoder, VAE and tokenizer come
|
|
|
|
| 34 |
| `primesim_base.safetensors` | the midtrained model -- start here to fine-tune |
|
| 35 |
| `primesim_posttrain_g2.safetensors` | stage 4, causal few-step model -- use for inference |
|
| 36 |
|
| 37 |
+
Run `primesim_posttrain_g2` on the demos in the
|
| 38 |
+
[repository](https://github.com/PrimebotTech/PrimeSim) to see what posttraining adds.
|
| 39 |
+
|
| 40 |
## Requirements
|
| 41 |
|
| 42 |
**These weights are not self-contained.** The text encoder, VAE and tokenizer come
|