14 lines
500 B
Markdown
14 lines
500 B
Markdown
|
---
|
||
|
|
license: mit
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
base_model:
|
||
|
|
- Qwen/Qwen3-8B
|
||
|
|
---
|
||
|
|
## RLM-Qwen3-8B-v0.1
|
||
|
|
|
||
|
|
A small, post-trained Qwen3-8B model from the experiments in the "Recursive Language Models" paper: [https://arxiv.org/abs/2512.24601](https://arxiv.org/abs/2512.24601).
|
||
|
|
|
||
|
|
The model was trained on trajectories produced using a fixed system prompt. It assumes the environment/scaffold from our RLM repo.
|
||
|
|
|
||
|
|
We recommend using vLLM with our inference code at https://github.com/alexzhang13/rlm to use it out of the box.
|