---
license: gemma
base_model: google/diffusiongemma-26B-A4B-it
base_model_relation: quantized
tags:
- gemma
- diffusion
- dllm
- block-diffusion
- int8
- w8a8
- compressed-tensors
- vllm
- ampere
pipeline_tag: text-generation
---

# DiffusionGemma 26B-A4B-it — INT8 W8A8 (dynamic)

INT8 weight + activation quantization of [google/diffusiongemma-26B-A4B-it](https://huggingface.co/google/diffusiongemma-26B-A4B-it), Google's block-diffusion language model (dLLM) built on the Gemma 4 26B-A4B MoE backbone.

**Why INT8 instead of FP8?** The existing [FP8-dynamic quantization](https://huggingface.co/RedHatAI/diffusiongemma-26B-A4B-it-FP8-dynamic) requires Hopper/Ada GPUs: on Ampere (sm_86, e.g. RTX 3090), vLLM's Marlin FP8 MoE kernel cannot tile this model's expert shape (K=352) within Ampere's 99 KB shared memory (`Invalid thread config`), and the Triton FP8 MoE backend rejects the per-channel × per-token scheme. **INT8 W8A8 routes through vLLM's Triton Int8 MoE backend, which runs on Ampere and newer** — this checkpoint serves on 2× RTX 3090.

## Quantization

- **Weights:** int8, symmetric, per-output-channel (RTN, `memoryless_minmax`)
- **Activations:** int8, symmetric, per-token, **dynamic** (computed at runtime, nothing stored)
- **Format:** `compressed-tensors`, `int-quantized`
- **Ignored (kept bf16):** `lm_head` (tied), embeddings, MoE routers, vision tower, self-conditioning — same ignore list as the FP8-dynamic release
- **Coverage:** 11,725 quantized weight tensors (all 128 experts × 48 layers split per-expert + dense MLP/attention projections), identical tensor layout to the FP8-dynamic release
- **Size:** ~26 GB (from ~49 GB bf16)

Quantization was performed by direct streaming RTN over the base safetensors (one shard at a time, CPU-only). Note for reproducers: the base checkpoint stores MoE experts **fused** (`experts.gate_up_proj` `[E, 2I, H]`, `experts.down_proj` `[E, H, I]`); this release splits them into per-expert `gate_proj`/`up_proj`/`down_proj` Linears so `compressed-tensors`/vLLM load them with `targets: [Linear]`. (llm-compressor `oneshot` with transformers ≥ 5.11 silently skips the fused 3-D expert tensors — 88% of parameters — producing a barely-compressed model.)

**Fidelity vs base weights:** relative reconstruction error < 1.2%, cosine similarity ≈ 0.99996 (experts and dense layers). Spot-checked outputs match base-model behavior, including its quirks (see below).

## Serving with vLLM

Requires a vLLM build with DiffusionGemma support (the `vllm/vllm-openai:gemma` image / diffusion branch; mainline ≤ 0.14-nightly does not include it).

```bash
vllm serve <this-repo> \
  --trust-remote-code \
  --max-num-seqs 4 \
  --hf-overrides '{"diffusion_sampler": "entropy_bound", "diffusion_entropy_bound": 0.1}' \
  --override-generation-config '{"max_new_tokens": 8192}' \
  --enable-auto-tool-choice \
  --reasoning-parser gemma4 \
  --tool-call-parser gemma4 \
  --default-chat-template-kwargs '{"enable_thinking": true}'
```

Optional faster decoding (~3× over the 48-step default, minor quality trade):

```bash
  --diffusion-config '{"canvas_length": 256, "max_denoising_steps": 16}' \
  --hf-overrides '{"diffusion_sampler": "entropy_bound", "diffusion_entropy_bound": 0.1, "diffusion_confidence_threshold": 0.0}'
```

### Operational notes (important)

- **Default `max_new_tokens` is 256** (from `generation_config.json`). With thinking enabled, reasoning will consume the whole budget and `content` comes back empty/truncated for clients that don't set `max_tokens` (e.g. OpenWebUI). Strongly recommended: `--override-generation-config '{"max_new_tokens": 8192}'` as above.
- **Reasoning output field:** with `--reasoning-parser gemma4`, parsed chain-of-thought is returned in `message.reasoning` (not `reasoning_content`) on current diffusion-branch builds.
- **Use the bundled `chat_template.jinja` as-is.** The model self-emits `<|channel>thought ... <channel|>` delimiters; pre-opening the channel in a custom template suppresses them.
- **Memory under concurrency:** each concurrent diffusion decode materializes ~1.9 GB of full-vocab fp32 sampler buffers *beyond* vLLM's profiled budget. Leave `--gpu-memory-utilization` headroom accordingly (e.g. 24 GB cards: ≈0.75 at `--max-num-seqs 4`).
- **Tensor parallel on multi-GPU consumer cards:** the diffusion-branch sampler assumes an unsharded embedding for its self-conditioning matmul (`probs @ embed_weight`) and its prompt-logprobs path; TP > 1 requires small patches until fixed upstream (all-reduce of vocab-sharded partials; disable prompt-logprobs). Single-GPU (≥ 32 GB) needs no patches. On PCIe-linked consumer cards also pass `--disable-custom-all-reduce`.

## Known model behaviors (present in the bf16 original — not quantization artifacts)

- Trivial imperative prompts ("Say OK.") can deterministically commit an empty canvas (immediate EOS, empty output). Verified identical on the bf16 base via the transformers reference implementation.
- Thinking consumes the output token budget; use `max_tokens ≥ 1024` for thinking-enabled requests.

## License

Gemma. This is a derivative (quantized) version of `google/diffusiongemma-26B-A4B-it`; use is subject to the [Gemma Terms of Use](https://ai.google.dev/gemma/terms).