Open-weight model
SimpleMemVLN-R2R-RxR15deg-Window8-Text-DualLane-FromBase
by Anh Dao anhdao69/SimpleMemVLN-R2R-RxR15deg-Window8-Text-DualLane-FromBase
SimpleMemVLN-R2R-RxR15deg-Window8-Text-DualLane-FromBase is an open-weight model from Anh Dao. Its published files total 20.8 GB.
Completed two-epoch joint R2R + English-guide RxR 15-degree training run. epoch-1 contains update 3,852; epoch-2 contains the final update 7,704. Epoch 1 is a mid-schedule snapshot; epoch 2 completes the planned schedule.
Model Card
Completed two-epoch joint R2R + English-guide RxR 15-degree training run. epoch-1 contains update 3,852; epoch-2 contains the final update 7,704. Epoch 1 is a mid-schedule snapshot; epoch 2 completes the planned schedule. No navigation evaluation result is claimed for these weights. Original Qwen3.5-4B base revision 851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a. https://github.com/anhdao69/SimpleMemVLN/tree/streamingtextdual. 30,815 episodes; 3,128,624 action steps per dataset pass before distributed tail padding. Two epochs, 7,704 updates; 232 warmup updates; seed 429. Four H100 GPUs, one complete episode per rank per microstep, gradient accumulation 2, effective global batch 8. BF16, ZeRO-2…
Excerpt from the card by Anh Dao.
Identity and Version
- Repository
- anhdao69/SimpleMemVLN-R2R-RxR15deg-Window8-Text-DualLane-FromBase
- Publisher
- Anh Dao
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- Not stated by the source
- Languages
- rxr
- Revision
- fa24de8b58db484d3986b49a62a6ce0c3190e1e0
- First published
- 2026-10-07
- Last updated
- 2026-10-08
Files and Weights
20 files, 20.8 GB in total. The weights are 2 files totalling 20.8 GB in bin.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| epoch-1/pytorch_model.bin | Weights | 10.4 GB | cd7b36d160c1 |
| epoch-2/pytorch_model.bin | Weights | 10.4 GB | 64f62f97c8a4 |
| SHA256SUMS.json | Configuration | 1.8 KB | — |
| epoch-1/config.json | Configuration | 2.9 KB | — |
| epoch-1/navigation.json | Configuration | 3.3 KB | — |
| epoch-1/processor_config.json | Configuration | 1.2 KB | — |
| epoch-1/trainer_state.json | Configuration | 719.0 KB | — |
| epoch-2/config.json | Configuration | 2.9 KB | — |
| epoch-2/navigation.json | Configuration | 3.3 KB | — |
| epoch-2/processor_config.json | Configuration | 1.2 KB | — |
| epoch-2/trainer_state.json | Configuration | 1.4 MB | — |
| resolved_config.json | Configuration | 2.0 KB | — |
| README.md | Documentation | 2.3 KB | — |
| epoch-1/chat_template.jinja | Other | 7.8 KB | — |
| epoch-2/chat_template.jinja | Other | 7.8 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| epoch-1/tokenizer.json | Tokenizer | 20.0 MB | 06b9509352d2 |
| epoch-1/tokenizer_config.json | Tokenizer | 1.2 KB | — |
| epoch-2/tokenizer.json | Tokenizer | 20.0 MB | 06b9509352d2 |
| epoch-2/tokenizer_config.json | Tokenizer | 1.2 KB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 20.8 GB
Released by Anh Dao through its official repository on Hugging Face.
Built From
- Derived from Qwen/Qwen3.5-4B
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 20.8 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.