> ## Documentation Index
> Fetch the complete documentation index at: https://miles.radixark.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# DeepSeek

> Miles recipes for the DeepSeek family — DeepSeek-V4 Flash (sparse-MLA + DSA indexer), V3, and R1.

Miles ships recipes for the DeepSeek family across two generations: **DeepSeek-V4 Flash** introduces sparse multi-head latent attention with a learned indexer and KV compressors (8-node H200), while **V3 / R1** remain the canonical 16-node 671 B-parameter recipes (BF16 train + 128×128 block-wise FP8 rollout, DeepEP, DAPO-style dynamic sampling).

## Variants

| Model             | Active / Total | HF ID                               | Recipe                                                  |
| ----------------- | -------------- | ----------------------------------- | ------------------------------------------------------- |
| DeepSeek-V4-Pro   | 49 B / 1.6 T   | TBA                                 | [deepseek-v4-pro](/docs/models/deepseek/deepseek-v4-pro)     |
| DeepSeek-V4-Flash | 13 B / 284 B   | `sgl-project/DeepSeek-V4-Flash-FP8` | [deepseek-v4-flash](/docs/models/deepseek/deepseek-v4-flash) |
| DeepSeek-V3       | 37 B / 671 B   | `deepseek-ai/DeepSeek-V3`           | [deepseek](/docs/models/deepseek/deepseek)                   |
| DeepSeek-R1       | 37 B / 671 B   | `deepseek-ai/DeepSeek-R1`           | [deepseek](/docs/models/deepseek/deepseek)                   |

A validated DeepSeek-V4-Pro recipe is not yet available — see [`radixark/miles#1046`](https://github.com/radixark/miles/issues/1046) for tracking.

## Fastest path to train

DeepSeek-V4-Flash needs 8 nodes of 8× H200 and the `radixark/miles:latest` image:

```bash theme={null}
cd /root/miles
python scripts/run_deepseek_v4.py full-train \
   --model-name DeepSeek-V4-Flash-FP8 \
   --num-nodes 8 --num-gpus-per-node 8
```

DeepSeek-R1 needs 16 nodes of 8× H100:

```bash theme={null}
cd /root/miles
bash scripts/run-deepseek-r1.sh              # full 16-node run
```

See the [DeepSeek-V4 Flash](/docs/models/deepseek/deepseek-v4-flash) page for the V4 architecture summary, parallelism layouts, and known workarounds; see the [DeepSeek R1 / V3](/docs/models/deepseek/deepseek) page for the V3 flow — FP8 → BF16 conversion, Megatron parallelism layout (TP8 / PP4 / EP32 / CP4), per-arg walkthrough, and the alternate Python launcher (`scripts/run_deepseek.py`).

## Pairs well with

* [PD Disaggregation](/docs/advanced/pd-disaggregation) — 671 B is where PD really earns its keep.
* [P2P Weight Transfer](/docs/advanced/p2p-weight-transfer) — amortize weight sync across ranks.
* [Fault Tolerance](/docs/advanced/fault-tolerance) — node failures are inevitable at 16-node scale.