---
license: apache-2.0
language:
- en
tags:
- 26b
- aeon
- aeon-7
- dflash
- draft-model
- drafter
- gemma
- gemma-4
- gemma4
- google
- speculative-decoding
base_model: AEON-7/supergemma4-26b-abliterated-multimodal-nvfp4
library_name: transformers
pipeline_tag: text-generation
---

# SuperGemma4-26B DFlash Draft (pilot / PoC)

This is a **proof-of-concept DFlash block-diffusion drafter** trained against
[AEON-7/supergemma4-26b-abliterated-multimodal-nvfp4](https://huggingface.co/AEON-7/supergemma4-26b-abliterated-multimodal-nvfp4)
(the NVFP4-quantized SuperGemma4 26B Abliterated Multimodal model, bf16 twin).

## What is this?

A small draft model used for **speculative decoding**: instead of the big 26B
target generating one token per step, the drafter proposes multiple tokens in
parallel using block diffusion, and the target verifies them in a single pass.
This can 2-3× generation throughput once the drafter is accurate enough.

**This release is a pilot** (5K samples, 1 epoch, ~28 min on 1× RTX PRO 6000
Blackwell). Top-1 match rate with the target is only ~5.8%, too low to actually
speed up generation. The artifact exists to validate the full training +
export + deployment pipeline end-to-end. See *Roadmap* below.

## Architecture

| Field | Value |
|---|---|
| Type | DFlash block-diffusion drafter (Qwen3-style) |
| Layers | 5 |
| Hidden | 2816 (matches target) |
| Heads | 22 attention / 22 KV |
| Head dim | 128 (vLLM-compatible) |
| Intermediate | 9728 |
| Vocab | 262144 |
| Max pos | 262144 |
| Block size | 8 |
| Target layer anchors | [1, 8, 14, 20, 27] |
| Parameters | ~570M |
| Dtype | bfloat16 |

## Training

| Field | Value |
|---|---|
| Target | SuperGemma4 26B (bf16, transformers 5.5.4 layout) |
| Data | `HuggingFaceH4/ultrachat_200k` train_sft, first 5000 conversations |
| Seq length | 2048 |
| Optimizer | AdamW (fused) |
| LR | 6e-4 (linear decay, 50 warmup steps) |
| Epochs | 1 |
| Steps | 5000 (bs=1, grad_accum=1) |
| Precision | bf16 mixed |
| Hardware | 1× NVIDIA RTX PRO 6000 Blackwell Server Edition (96 GB) |
| Wall time | 27:48 |
| Loss (final step) | ~5.9 (noisy; distillation + self-logit) |
| Train acc @ step 5000 | 5.79% (parallel_0_step_0 top-1) |

## Training infra notes

Built with [NVIDIA/TensorRT-Model-Optimizer#1211](https://github.com/NVIDIA/TensorRT-Model-Optimizer/pull/1211)
(DFlash training mode, merged in `modelopt` 0.43.0rc2.dev). Required two
local patches:

1. `EagleTrainerWithAccLog._save` override to save **drafter-only** (1.2 GB)
   instead of the full 50 GB frozen target + drafter tree — stock HF Trainer
   would OOM the filesystem.
2. `modeling_gemma4.py` line 2027 — bypass the
   `mm_token_type_ids required when training` check for text-only training.

## Deployment with vLLM

```bash
vllm serve AEON-7/supergemma4-26b-abliterated-multimodal-nvfp4 \
    --speculative-config '{
        "method":"draft_model",
        "model":"AEON-7/supergemma4-26b-dflash-pilot",
        "num_speculative_tokens":5
    }' \
    --quantization modelopt \
    --max-model-len 65536
```

⚠️  With only 5.8% top-1, expect **negative** speedup (verification overhead
swamps any gain). This pilot is for *plumbing validation*, not perf.

## Roadmap

This pilot proves the training stack. To make it actually fast:

| Stage | Data | Epochs | Expected top-1 | ETA |
|---|---|---|---|---|
| **Pilot (this)** | 5K | 1 | 5.8% | 28 min × 1 GPU |
| Small | 50K | 3 | ~25% | ~15 hr × 1 GPU |
| Medium | 500K | 3 | ~55% | ~6 days × 1 GPU (or 18 hr × 8 GPUs) |
| Production | 2M | 5-10 | 70-80% | ~1 week × 8 GPUs |

Production-quality domain-general data (mix of ShareGPT, UltraChat,
Magpie, LMSYS Chat, code) rather than UltraChat alone is also key.

## Files

- `model.safetensors` — 1.2 GB, 58 tensors (5-layer drafter)
- `config.json` — Qwen3-style DFlashDraftModel config
- `tokenizer.json`, `tokenizer_config.json`, `chat_template.jinja` — from target
- `trainer_state.json` — loss/acc history per step
- `training_args.bin` — exact training config for reproducibility

## License

Apache 2.0. This is derived work from the SuperGemma4 26B target.

---

## ☕ Support the work

If this release has been useful, tips are deeply appreciated — they go directly toward more compute, more models, and more open releases.

<table align="left">
  <tr><td align="left">
    <strong>₿ Bitcoin (BTC)</strong><br/>
    <img src="https://raw.githubusercontent.com/AEON-7/AEON-7/main/assets/qr/btc.png" alt="QR" width="200"/><br/>
    <sub><code>bc1q09xmzn00q4z3c5raene0f3pzn9d9pvawfm0py4</code></sub>
  </td></tr>
  <tr><td align="left">
    <strong>Ξ Ethereum (ETH)</strong><br/>
    <img src="https://raw.githubusercontent.com/AEON-7/AEON-7/main/assets/qr/eth.png" alt="QR" width="200"/><br/>
    <sub><code>0x1512667F6D61454ad531d2E45C0a5d1fd82D0500</code></sub>
  </td></tr>
  <tr><td align="left">
    <strong>◎ Solana (SOL)</strong><br/>
    <img src="https://raw.githubusercontent.com/AEON-7/AEON-7/main/assets/qr/sol.png" alt="QR" width="200"/><br/>
    <sub><code>DgQsjHdAnT5PNLQTNpJdpLS3tYGpVcsHQCkpoiAKsw8t</code></sub>
  </td></tr>
  <tr><td align="left">
    <strong>ⓜ Monero (XMR)</strong><br/>
    <img src="https://raw.githubusercontent.com/AEON-7/AEON-7/main/assets/qr/xmr.png" alt="QR" width="200"/><br/>
    <sub><code>836XrSKw4R76vNi3QPJ5Fa9ugcyvE2cWmKSPv3AhpTNNKvqP8v5ba9JRL4Vh7UnFNjDz3E2GXZDVVenu3rkZaNdUFhjAvgd</code></sub>
  </td></tr>
</table>

> **Ethereum L2s (Base, Arbitrum, Optimism, Polygon, etc.) and EVM-compatible tokens** can be sent to the same Ethereum address.