---

license: other

license_name: qwen-community-license-1.0

license_link: https://huggingface.co/Qwen/Qwen3.8-Flash-Next/blob/main/LICENSE

base_model: Qwen/Qwen3.8-Flash-Next

tags:

• gguf
• speculative-decoding
• mtp
• qwen4exp
• xyntetik
• xyntetik-runner

---

Qwen3.8-Flash-Next MTP head, GGUF companion

What this is and is not. This is the multi-token-prediction (MTP, "NextN") draft head of

Qwen3.8-Flash-Next as one GGUF file, built from the publisher's BF16 checkpoint. It is not a

model and it does not run on its own: it is loaded beside the trunk GGUF and proposes draft

tokens that the trunk verifies, so the generated text is the trunk's and only the speed

changes. It was built for the ISTA-DASLab GSQ-RCO IQ3_S trunk

(ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF,

part 1 sha256 recorded in the evidence below), which ships without the head. It is a

draft head of the base model's weights, no fine-tuning, no calibration.

| File | Size | sha256 |

|---|---|---|

| Qwen3.8-Flash-Next-MTP-Q8_0.gguf | 4,165,605,056 bytes | cd2f837a63f0b6ea1d951976e161fc40648932ffaf59194aed3557175638000b |

Measured

Same box, same trunk file, same prompts, greedy, 128 new tokens, two runs each

(Xyntetik Strata one-to-one study, 2026-10-07; engine-reported figures).

| Engine | Setting | 512-token prompt decode | 4K prompt decode | Drafts accepted |

|---|---|---|---|---|

| llama.cpp b11433, CPU only, 24 threads | plain | 13.1, 13.2 tok/s | 12.5, 12.9 tok/s | |

| llama.cpp b11433, CPU only, 24 threads | --spec-type draft-mtp with this file | 18.3, 20.4 tok/s | 18.9, 17.0 tok/s | 69-82% |

On the Blackwell MIG 1g.24gb slice (CUDA, --fit on -c 8192 --parallel 2, the same launch as the plain rows):

| llama.cpp b11433, CUDA | plain | with this head | Drafts accepted |

|---|---|---|---|

| 512-token prompt, decode tok/s | 25.5, 27.2 | 39.0, 34.9 | 67-72% |

| 4K prompt, decode tok/s | 26.7, 25.2 | 37.1, 38.9 | 61-69% |

| two concurrent 512-token requests, decode tok/s each | 20.5 and 20.5 | 26.3 and 25.7 | 69-74% |

| prefill tok/s | 151-222 (512), 285 (4K) | 227-229 (512), 275-279 (4K) | |

The full tables are in the study's report, linked from the Runner repository when it lands.

Known from the same measurements: in llama.cpp the drafted run is not byte-identical to the

plain run (the 4K prompt's text diverged once). In Xyntetik Runner 1.1.1 the head runs on the

CPU and on CUDA (--mtp-file), drafts accepted 4.1 to 5.3 of 5 per round, and the greedy text

is byte-identical to plain decoding within one expert split; on a 24 GB device split of this

model it is still slower than plain decoding (IQ3_S: 12.0 against 13.8 tok/s on the same split,

CPU only: 7.7 against 9.3) because the verify tile runs 23 to 30 expert layers on the host one

row at a time and a tile's rows share no expert at 10-of-512 routing. The head pays where the

device holds the trunk; on this slice treat it as a correctness feature. Measured 2026-10-08,

Runner 1.1.1, one MIG 1g.24gb slice; the head's block takes the VRAM of one expert layer under

--cpu-moe auto, so compare against a baseline pinned to the same --cpu-moe N.

Use

llama.cpp (b11433 or newer with qwen4exp MTP support):

llama-server -m Qwen3.8-Flash-Next-GSQ-RCO-IQ3_S-00001-of-00002.gguf \
  -md Qwen3.8-Flash-Next-MTP-Q8_0.gguf --spec-type draft-mtp --fit on

Xyntetik Runner 1.1.1 or newer:

runner -m Qwen3.8-Flash-Next-GSQ-RCO-IQ3_S-00001-of-00002.gguf \
  --mtp-file Qwen3.8-Flash-Next-MTP-Q8_0.gguf --gpu off

How it was built

scripts/make-mtp-companion.py in the Xyntetik Runner

repository, from 33 tensors of Qwen/Qwen3.8-Flash-Next at revision

de4b8e4d43b917e7706784d8bb445c9af86a3540 fetched by HTTP byte range: the 31 mtp.* tensors,

model.language_model.embed_tokens.weight and lm_head.weight (so the file also loads alone as

a draft model). The layout is llama.cpp's MTP-only export for this architecture: block_count

49 with nextn_predict_layers 1, the block at blk.48., the head at blk.48.nextn.,

eh_proj = [fc_embedding | fc_hidden], the fused gate_up_proj split gate-first, every RMS-norm

gamma stored +1 as the converter does for this family, the hyperparameters and tokenizer copied

from the trunk's header byte for byte. Storage: norms F32; hyper-connection, router and indexer

tensors BF16 as published; attention, expert, shared-expert, embedding and LM-head matrices Q8_0.

The header carries the provenance under xyntetik.mtp.*: source repo, revision, the 33 source

tensor names and their sha256 as fetched, and the builder.

The head's acceptance rate on the real model (69-82% of drafts accepted, in line with the

publisher's own MTP figures) is the check that the construction is right; a wrong head drafts

nothing useful.

The trunk with the head in one set (trunk-with-head/)

For anyone who wants one model directory instead of a trunk from one repository and a head from

another: trunk-with-head/ holds a 4-part split GGUF that is ISTA-DASLab's IQ3_S trunk with its

header extended for the head (block_count 49, nextn_predict_layers 1, one more compress ratio, the

split counts) and this head as the last part; the trunk's tensor bytes are untouched (its 54.8 GB first

file is split in two for the hub's 50 GB file limit). Tested: Runner 1.1.0 --mtp on part 1 drafts

from the head inside the set, text byte-identical to plain decoding and to the original trunk;

llama.cpp b11433 loads it with --spec-type draft-mtp and no -md (62 of 99 drafts accepted on

the CPU-only check). Part sha256 in trunk-with-head/SHA256SUMS. The trunk is ISTA-DASLab's

work (Apache-2.0); the whole set is a derivative under the Qwen Community License 1.0.

llama-server -m trunk-with-head/Qwen3.8-Flash-Next-GSQ-RCO-IQ3_S-with-MTP-head-00001-of-00004.gguf --spec-type draft-mtp --fit on
runner -m trunk-with-head/Qwen3.8-Flash-Next-GSQ-RCO-IQ3_S-with-MTP-head-00001-of-00004.gguf --mtp --gpu off

License and attribution

The weights are Qwen's, redistributed as a derivative under the

Qwen Community License 1.0

with this notice: built from Qwen/Qwen3.8-Flash-Next by Xyntetik (Zenova AB), 2026-10-07.

The trunk this companion was built for is ISTA-DASLab's quantization, published under

Apache-2.0 by its authors; it is not included here.

<!-- xyntetik-program -->

About Xyntetik

This repository is published by Xyntetik, the Zenova AB program for running capable language models on hardware you own and proving what they did. Its released engine is Xyntetik Runner: a free, single-binary GGUF engine in C for CPU, CUDA and Metal that serves, scores and trains through the quantized file it deploys and records every run as a receipt. The measurements behind the program's claims are collected at xyntetik.com/evidence. The program's organization on Hugging Face is Xyntetik.