193 lines
8.0 KiB
Markdown
193 lines
8.0 KiB
Markdown
# DFlash and DSpark Speculative Decoding
|
|
|
|
## Overview
|
|
|
|
DFlash and DSpark are the two newest speculative decoding methods. They draft many tokens in one parallel pass instead of one token at a time. This reduces decode latency on memory-bound workloads.
|
|
|
|
DFlash comes from Z Lab at UC Berkeley. DSpark comes from DeepSeek. Both work with vLLM and llama.cpp. Both have draft models for the Qwen3.6 models that Dankserver and Dankclaw run.
|
|
|
|
## How DFlash Works
|
|
|
|
DFlash is a lightweight block diffusion model. It predicts a block of tokens in a single forward pass. Key design points:
|
|
|
|
- The draft model reuses the embedding and LM head from the target model.
|
|
- Only the intermediate layers are trained.
|
|
- It fuses target-model hidden features into the KV projections of every draft layer.
|
|
- A 16-token DFlash drafter has lower latency than a 1-layer EAGLE-3 drafter producing 8 tokens.
|
|
|
|
Results from the DFlash paper (arXiv:2602.06036):
|
|
|
|
- Over 6x lossless speedup on Qwen3-8B.
|
|
- Up to 2.5x higher speedup than EAGLE-3.
|
|
- Diffusion drafting cost is flat, so block size does not increase cost.
|
|
|
|
## How DSpark Works
|
|
|
|
DSpark is a semi-autoregressive extension of DFlash from DeepSeek (arXiv:2607.05147). It keeps the DFlash parallel diffusion backbone and adds a lightweight sequential Markov head. This models inter-token dependencies inside each block. It also uses confidence-scheduled verification, which stops verifying tokens that have a high rejection risk.
|
|
|
|
Results from the DSpark paper:
|
|
|
|
- 60-85% faster per-user generation than MTP-1 on DeepSeek-V4 production traffic.
|
|
- Useful when parallel drafts decay in acceptance rate.
|
|
|
|
Both DFlash and DSpark are lossless under rejection sampling. Use the strict rejection sampler.
|
|
|
|
## Runtime Support
|
|
|
|
| Runtime | DFlash | DSpark | MTP |
|
|
|---------|--------|--------|-----|
|
|
| vLLM | yes (v0.20.1+) | yes | yes |
|
|
| llama.cpp | yes | yes | yes |
|
|
| SGLang | yes | no | yes |
|
|
|
|
### vLLM
|
|
|
|
vLLM v0.20.1+ has core DFlash support. Use the `--speculative-config` JSON object.
|
|
|
|
DFlash on vLLM:
|
|
|
|
```
|
|
vllm serve Qwen/Qwen3.6-35B-A3B \
|
|
--speculative-config '{"method": "dflash", "model": "z-lab/Qwen3.6-35B-A3B-DFlash", "num_speculative_tokens": 15}' \
|
|
--attention-backend flash_attn \
|
|
--max-num-batched-tokens 32768
|
|
```
|
|
|
|
Legacy flags also work:
|
|
|
|
```
|
|
vllm serve Qwen/Qwen3.6-35B-A3B \
|
|
--spec-method dflash \
|
|
--spec-model z-lab/Qwen3.6-35B-A3B-DFlash \
|
|
--spec-tokens 8
|
|
```
|
|
|
|
DFlash requires a non-causal attention backend. Set `"attention_backend": "flash_attn"` in the speculative config if needed. On ROCm, check that the flash_attn backend is available before using DFlash.
|
|
|
|
DSpark on vLLM:
|
|
|
|
```
|
|
vllm serve <target-model> \
|
|
--speculative-config '{"method": "dspark", "model": "<dspark-draft>", "num_speculative_tokens": 15}'
|
|
```
|
|
|
|
### llama.cpp
|
|
|
|
llama.cpp supports DFlash, DSpark, and MTP through `--spec-type`. When you pass `--spec-draft-hf`, llama.cpp auto-discovers the sidecar files and infers the type. The priority order is DSpark, DFlash, EAGLE-3, MTP.
|
|
|
|
`--spec-type` accepts a **comma-separated list** of speculative implementations. `common_speculative_init` (in `common/speculative.cpp`) builds a vector of implementations and chains drafting through all of them (`common_speculative_draft_params.drafting` is "used to chain the drafts through all the available implementations"). This means you can **layer ngram-mod on top of a DFlash draft model** in one server process:
|
|
|
|
```
|
|
llama-server --hf-repo <target-GGUF> \
|
|
--spec-draft-hf <dflash-draft-GGUF> \
|
|
--spec-type draft-dflash,ngram-mod
|
|
```
|
|
|
|
The DFlash drafter proposes, then ngram-mod runs as an additional implementation in the chain. Verified against upstream `common/speculative.cpp` (b10133+).
|
|
|
|
DFlash on llama.cpp:
|
|
|
|
```
|
|
llama-server --hf-repo <target-GGUF> \
|
|
--spec-draft-hf <draft-GGUF> \
|
|
--spec-type draft-dflash \
|
|
--spec-draft-n-max 16
|
|
```
|
|
|
|
DSpark on llama.cpp:
|
|
|
|
```
|
|
llama-server --hf-repo <target-GGUF> \
|
|
--spec-draft-hf <draft-GGUF> \
|
|
--spec-type draft-dspark
|
|
```
|
|
|
|
MTP on llama.cpp:
|
|
|
|
```
|
|
llama-server --hf-repo <target-with-MTP> \
|
|
--spec-type draft-mtp \
|
|
--spec-draft-n-max 2
|
|
```
|
|
|
|
## Draft Models
|
|
|
|
DFlash draft model sizes for the models on these servers:
|
|
|
|
| Target model | DFlash draft | Draft size |
|
|
|--------------|--------------|------------|
|
|
| Qwen3.6-35B-A3B | z-lab/Qwen3.6-35B-A3B-DFlash | 0.4B |
|
|
| Qwen3.6-27B | z-lab/Qwen3.6-27B-DFlash | 2B |
|
|
| Qwen3.5-35B-A3B | z-lab/Qwen3.5-35B-A3B-DFlash | 0.4B |
|
|
| Qwen3.5-27B | z-lab/Qwen3.5-27B-DFlash | 2B |
|
|
|
|
The DFlash draft is not a standalone model. It must pair with its target model. The model cards recommend block size 8 for higher concurrency and 16 for single-stream throughput.
|
|
|
|
## Deployment on Dankserver (NVIDIA RTX 3090)
|
|
|
|
## Deployment on Dankserver (NVIDIA RTX 3090)
|
|
|
|
The active Dankserver deployment runs **llama.cpp** (`frank-default-small`), not vLLM. It serves the HauhauCS Qwen3.6-35B-A3B uncensored `Q6_K_P` on port 7770 using **ngram-mod** speculative decoding (n_min=48 / n_max=64 / n_match=24) with 3 slots at 200000 context each and MoE experts on CPU. See [[deployment]] and `scripts/serve-llama-frank-small.sh`.
|
|
|
|
> **DFlash retired from active service.** `DFlash` (draft-dflash) was tested initially but did not deliver the desired speed gains under the `frank-default-small` workload (see ngram-mod benchmarks below). The `Alittlehammmer/Qwen3.6-35B-A3B-DFlash` draft GGUF remains in `/home/dank/models/` as reference material only.
|
|
|
|
The vLLM material below documents the archived AWQ/DFlash path on the same hardware. The RTX 3090 is Ampere (compute capability 8.6). Use these conditions:
|
|
|
|
- FP8 KV cache is valid on CUDA 12.
|
|
- FP8 weights compute through a fallback path; no native FP8 tensor cores.
|
|
- The DFlash draft model (0.4B) fits easily in 24 GB VRAM alongside the target.
|
|
|
|
Working DFlash flags for this server:
|
|
|
|
```
|
|
--spec-method dflash
|
|
--spec-model z-lab/Qwen3.6-35B-A3B-DFlash
|
|
--spec-tokens 8
|
|
--kv-cache-dtype fp8
|
|
```
|
|
|
|
Example DFlash + AWQ on the RTX 3090:
|
|
|
|
```
|
|
docker run --rm --name vllm-qwen36-dflash \
|
|
--gpus all --ipc=host \
|
|
-p 8000:8000 \
|
|
-v ~/.cache/huggingface:/root/.cache/huggingface \
|
|
docker.io/vllm/vllm-openai:latest \
|
|
cyankiwi/Qwen3.6-35B-A3B-AWQ-4bit \
|
|
--speculative-config '{"method": "dflash", "model": "z-lab/Qwen3.6-35B-A3B-DFlash", "num_speculative_tokens": 8}' \
|
|
--attention-backend flash_attn \
|
|
--gpu-memory-utilization 0.95
|
|
```
|
|
|
|
## Deployment on Dankclaw (AMD Strix Halo gfx1151)
|
|
|
|
Dankclaw now runs llama.cpp as its runtime for `frank-default-large` (port 7770), serving the HauhauCS Qwen3.6-27B uncensored `Q8_K_P` model with a **DFlash draft model** (`draft-dflash`) on the ROCm (HIP) backend. The earlier vLLM-ROCm CIRU INT4 Ling-3.0-flash runtime (native MTP) was retired because the model had problems.
|
|
|
|
Current facts:
|
|
|
|
- Main model: HauhauCS/Qwen3.6-27B-Uncensored-HauhauCS-Aggressive `Q8_K_P` (dense 27B).
|
|
- Draft model: Alittlehammmer/Qwen3.6-27B-DFlash-GGUF-llama.cpp `Q8_0` (block size 16).
|
|
- Spec command: `--spec-type draft-dflash` with `--spec-draft-n-max 8`.
|
|
- Backend `--device ROCm0`; `--gpu-layers all --fit off`; KV `q8_0` K / `f16` V.
|
|
- Context: 3 x 200700 = 602100; batch 2048 / ubatch 512.
|
|
|
|
Benchmark result: ROCm beats Vulkan, and DFlash-alone generates faster than a DFlash+ngram layer at 4k. See [[frank-default-large-benchmark]] for full numbers and the memory-bound batch/ubatch limits.
|
|
|
|
## Notes
|
|
|
|
- Speculative decoding helps low-QPS, memory-bound, latency-focused workloads. It helps less at high QPS.
|
|
- DFlash needs a non-causal attention backend; the ROCm flash_attn backend must be validated before production use.
|
|
- The old `--swap-space` flag no longer exists in current vLLM. Use `--kv-offloading-size`.
|
|
- `--speculative-config` is the preferred CLI. Legacy `--spec-method`, `--spec-model`, and `--spec-tokens` flags still exist.
|
|
- Use `rejection_sample_method` default `strict` for lossless output.
|
|
|
|
## See Also
|
|
|
|
- [[Quantization by Hardware Architecture]]
|
|
- [[Running Qwen3.6 35B A3B FP8 on Dankserver]]
|
|
- [[Running Qwen3.6 27B FP8 on Dankclaw]]
|
|
- [[Dankclaw System Specs]]
|
|
- [[Dankserver System Specs]]
|
|
- [[VLLM Overview]]
|
|
- [[deployment]] |